88093
          Republic of India
Accelerating Agricultural
    Productivity Growth
                  May 21, 2014
Standard Disclaimer
This volume is a product of the staff of the International Bank for Reconstruction and
Development/The World Bank. The findings, interpretations, and conclusions expressed in this
paper do not necessarily reflect the views of the Executive Directors of The World Bank or the
governments they represent. The World Bank does not guarantee the accuracy of the data included
in this work. The boundaries, colors, denominations, and other information shown on any map in
this work do not imply any judgment on the part of The World Bank concerning the legal status of
any territory or the endorsement or acceptance of such boundaries.




Copyright Statement
The material in this publication is copyrighted. Copying and/or transmitting portions or all of
this work without permission may be a violation of applicable law. The International Bank for
Reconstruction and Development/The World Bank encourages dissemination of its work and will
normally grant permission to reproduce portions of the work promptly.

For permission to photocopy or reprint any part of this work, please send a request with complete
information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA,
telephone: 978-750-8400, fax: 978-750-4470, http://www.copyright.com/.

All other queries on rights and licenses, including subsidiary rights, should be addressed to
the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA,
fax: 202-522-2422, e-mail: pubrights@worldbank.org.




Design and Print: Macro Graphics Pvt. Ltd. www.macrographics.com
                 Contents




Preface                                                                    xi
Acknowledgments                                                       xiii
Acronyms and Abbreviations                                                 xv
Overview                                                             xvii

PART A: CONTEXT AND OBJECTIVES OF THIS REPORT	                              1
Chapter 1: Introduction to the Issues and Analytical Approach               1
 Motivation and Scope of this Study                                         4
 Specific Objectives and Analytical Approach                                5
 The Broader Context and Structure of This Report                           6

        Performance of Indian Agriculture: Temporal, Spatial, and 
PART B:                                                                     7
        Comparative Perspectives
Chapter 2: The Context and Conundrums of Contemporary Agriculture           7
 India’s Food Security Puzzle                                               7
 Persistent Food Price Rise                                                 8
 Trade Policy and Cereal Prices                                            10
 Food Grain Management and Food Prices                                     11
 Rising Real Rural Wages                                                   12
 Persistent Questions about the Sustainability of Growth                   13
Chapter 3: Temporal Trends in Agricultural Output and Productivity         15
 Evolution of Agricultural Growth                                          15
 Why Did Growth Stagnate from 1997 to 2005?                                18
 Implications of Findings                                                  26




                                                                      iii
Republic of India: Accelerating Agricultural Productivity Growth




Chapter 4: Spatial Heterogeneity: Performance at the Subnational Level              29
 Agricultural Growth Across States                                                  29
 Productivity Changes at the District Level                                         31
 Dynamics of Agricultural Productivity at the District Level                        33
 Implications of the Subnational Analysis                                           41
Chapter 5: StructuraL Change                                                        43
 India’s Structural Transformation in an International Perspective                  44
 Drivers of Growth in the Comparator Countries: Implications for India              49
 Change at the Micro Level: Insights from Village Studies                           50
 Implications of Observations at the Micro Level                                    51

PART C: Sources of Agricultural Growth	                                             53
Chapter 6: Growth Decomposition                                                     55
 Changes in Subsector Contributions                                                 55
 Components Driving the Value of Crop Production                                    56
 Summary of Findings                                                                58
Chapter 7: Total Factor Productivity Growth                                         61
 Drivers of Productivity Growth at the National Level                               62
 Tfp at the State Level                                                             68
 Implications: Compromising Sustainability                                          72
Chapter 8: Productivity and Efficiency at the Household (Micro) Level               73
 Drivers of Changes in Productivity: Nonparametric Analysis                         73
 Changes in Efficiency Over Time: Parametric Analysis                               77
 Determinants of Productivity and Technical Efficiency                              81
 Changes in Agricultural Relations: Are Smaller Farms still Efficient and Viable?   83
 Summary and Implications: High Cost of Inefficiency                                85
Chapter 9: Technology: Yield Gaps and Prospects for Growth                          87
 Current Yield Gaps and Progress in Closing Gaps Over Time                          88
 Technological Progress: Evolution of Realizable Yields                             91
 Metrics of Research Achievement Other than Yield                                   92
 Technology Capital: Investing for the Future                                       93
 Conclusions and Implications                                                       97



  iv
                                                                                    ﻿Contents




Chapter 10: Livestock Subsector: Opportunities for Action to Improve Performance            99
 Introduction                                                                               99
 Opportunities for the Livestock Subsector                                              102
 Constraints and Challenges                                                             103
 Conclusions and Implications                                                           112

PART D: MOVING BEYOND THE FARM: INVESTMENTS, MARKETS AND FOOD PROCESSING 115
Chapter 11: Investments for Growth and Sustainability                                   117
 Trends in Public Expenditure                                                           118
 Trends in Private Investment                                                           120
 Impact of Expenditure Patterns: Sustaining Productivity                                122
 Conclusions and Implications                                                           126
Chapter 12: Current State of Agricultural Markets in India                              127
 Alternative Marketing Arrangements: Linking Smallholders to Markets                    129
 Efficacy of Market Channels in Improving Access                                        133
 Status of Agricultural Marketing Reforms                                               135
 Conclusions and Implications                                                           140
Chapter 13: Agro-Industry: The Food-Processing Sector                                   143
 Food Industry Structure and Investment                                                 144
 Patterns and Drivers of Private Investment in Food Manufacturing                       150
 Productivity Growth, Technical Change, and Technical Efficiency in the
 Food-Processing Industry                                                               153
 Summary and Implications                                                               157

PART E: Transforming Agriculture in East India                                          159
Chapter 14: Priorities for Agricultural Growth in Bihar and Odisha                      161
 Agricultural Performance                                                               162
 Challenges                                                                             165
 Priority Areas for Growth                                                              169
References                                                                              173
 Background Papers for this Report                                                      173
 Other References                                                                       174




                                                                                         v
Republic of India: Accelerating Agricultural Productivity Growth




Annexes                                                                                  189
 Annex 1: Policy Reforms and Private Sector Response                                     189
 Annex 2: Land Use, Grain Stocks and the Structure of Production                         193
 Annex 3: Comparing the Impact of Alternative Rainfall Estimates                         195
 Annex 4: Sub-National Heterogeneity in Agricultural Productivity                        197
 Annex 5: Changing Structure of Household Incomes in Bihar                               205
 Annex 6: Growth Decomposition: Methodology and Selected Indicators                      206
 Annex 7: Tfp: Methodology And Crop-Specific Estimates                                   211
 Annex 8: Changes In Farm Revenues And Efficiency By Land Size                           216
 Annex 9: Progress and Potential For Gains Through Technology Capital                    217
 Annex 10: Livestock Sector: Status And Performance                                      227
 Annex 11: Public Expenditure Patterns And Impact                                        231
 Annex 12: Status of Agricultural Market Reforms                                         234
 Annex 13: Food Processing: Structure and Performance                                    242
 Annex 14: Bihar and Odisha: Agriculture Performance and Constraints                     253
Tables
 Table 1: Sectoral Trend Growth Rates by Period (%/yr)                                    17
 Table 2: Trend Growth in Net State Domestic Product, 2004/05–2012/13                     30
 Table 3: District Growth Typologies                                                      34
 Table 4: District Growth Typologies by State (Long Run, 1970–2007)                       36
 Table 5: Levels and Changes in Key Indicators of Changes in Agricultural Productivity    40
 Table 6: Turning Points for Biic                                                         45
 Table 7: Distribution of Income by Source and Caste, 2011 (%)                            51
 Table 8: Growth in Crop Yields                                                           56
 Table 9: Tfp Estimates at the National, Sector-Wide Level                                63
            gricultural Sector Growth Accounts by Growth Episode, 1980–2008
 Table 10: A
           (trend growth rates)                                                           65
 Table 11: Decomposing Tfp Growth                                                         66
 Table 12: Tfp Estimates by State, Traditional Crops and Sector-Wide, 1998–2008           69
 Table 13: Regional Classification                                                        74
 Table 14: Average Technical Efficiency at the Regional and Country Level                 74
 Table 15: Tfp Growth Decomposition (Average Growth Rate, 1999–2007)                      75



  vi
                                                                                           ﻿Contents




 Table 16: Proportion of Profit Lost in Relation to Short-Run Potential Profit                       76
 Table 17: Rural and Urban Household Expenditure Patterns have Shifted Away from Cereals             99
 Table 18: Distribution of Land and Livestock Holding by Land Holding Size                      101
 Table 19: Employment in the Livestock Subsector by Farm Size Category                          101
 Table 20: Public Spending in the Livestock Subsector in India, Selected Years                  104
            overnment Expenditures on Agriculture and Allied Sectors
 Table 21: G
           (All India, 2004/05 Prices)                                                          118
 Table 22: On-Farm Investment (Capital per Acre), Real Rs. 1999                                 122
 Table 23: Comparison of Charges in Private and Apmc Markets                                    139
 Table 24: Price Build up in Tomato Value Chain (Salarpur, Uttar Pradesh)                       139
 Table 25: Organized Food-Processing Segment: Size and Growth                                   147
 Table 26: Unorganized Food-Processing Segment: Size and Growth                                 148
 Table 27: Classification of States as per LQ and Rdi Estimates Based on Investment             151
 Table 28: Milk Yields in Bihar and Odisha (kg/Day/per Animal) in 2009–10                       164
 Table 29: Fertilizer use in Bihar and Odisha                                                   166
            arm Harvest Prices (Fhp) of Paddy, Wheat, and Maize in Bihar and their
 Table 30: F
           Minimum Support Price (Msp) (Rs/Quintal)                                             167
Figures
 Figure 1: Per Capita Availability and Production of Cereals (grams/day)                              7
 Figure 2: Percentage Change in per Capita Consumption (kg/l/no.),1983/84–2009/10                     8
 Figure 3: Food Prices Relative to Prices of All Commodities                                          8
 Figure 4: Contribution to Food Price Rise (Percent)                                                  9
 Figure 5: Real Minimum Support Prices                                                                9
            elationship between Domestic and External Wheat Prices in Surplus Producing
 Figure 6: R
           Areas, 2005 Rs/MT (Left Panel) and Distortion in Agricultural Incentives (Right Panel) 10
 Figure 7: Cereal Trade, Procurement, and Food Price Index, 1988–2012                                12
 Figure 8: Agricultural Growth Rate, 1952–2012                                                       15
            emporal Performance of IndIan Agriculture Gdp: Trend Growth Rate and Growth
 Figure 9: T
           Volatility for Decades Ending 1960/61 to 2012/13                                          16
 Figure 10: Gross Fixed Capital Formation in Agriculture                                             18
 Figure 11: Agriculture Terms of Trade                                                               19
 Figure 12: Relationship between Agricultural Productivity and Rainfall                              21
 Figure 13: Distribution of Rainfall Shocks Across Zones, 2000–07                                    22



                                                                                                vii
Republic of India: Accelerating Agricultural Productivity Growth




 Figure 14: Growth in Irrigated Area by Source                                              22
 Figure 15: Historical Rainfall Patterns, 1901–2010                                         23
 Figure 16: Distribution of Rainfall Anomalies Across Districts and Years                   25
             en-Year Average Agricultural Gdp Growth Rates
 Figure 17: T
            (Actual and Counterfactual Scenarios)                                           26
 Figure 18: Factors Explaining Variation in Productivity of Cereal Crops                    29
 Figure 19: State Agriculture Growth Rates by Time Period, 1980/81–2011/12                  30
 Figure 20: Real Value of Agricultural Output per Hectare, 2005–07                          31
 Figure 21: Productivity Changes by District, 1990–2007                                     31
 Figure 22: Changes in Inter-District Equality (Gini Coefficient), 1991–2007                32
 Figure 23: Distribution of Yields Across Districts by Decade                               34
 Figure 24: Long-Run Productivity Typology of Districts Based on Cereal Yields              35
 Figure 25: Dispersion of Districts by Agricultural Productivity Typology                   35
 Figure 26: Share of Cereal Area Planted to High-Yielding Varieties                         36
 Figure 27: Irrigation Intensity Across District Typologies by Decade                       37
 Figure 28: Fertilizer Intensity Across District Typologies by Decade                       37
 Figure 29: Changes in District Agricultural Productivity                                   39
 Figure 30: Ratio of Value Added per Worker (Biic), Nonagriculture/Agriculture: 1980–2009   46
             hare of Value Added Minus Share of Employment in Agriculture,
 Figure 31: S
            Residuals (Biic): 1980–2009                                                     46
 Figure 32: Decomposition of Output Growth, Biic, Over the Long and Short Run               47
 Figure 33: Tfp Growth in Biic and Selected Regions                                         48
 Figure 34: Drivers of Tfp, India and China, 1961–2006                                      48
 Figure 35: The World Economic Forum’s Global Competitiveness Index                         49
 Figure 36: Contribution of Subsectors to Growth in Agricultural Gdp                        55
 Figure 37: Shares of Crops in Sectoral Growth                                              55
 Figure 38: Share of Value Divided by Share of Area by Crop Type                            56
 Figure 39: Decomposition of Growth in the Crops Sector                                     57
 Figure 40: Sources of Agricultural Growth by Region                                        59
 Figure 41: Contribution of Inputs and Tfp to Output Growth                                 64
             roductivity Change in Agriculture: Technical Progress Versus “Catching Up”
 Figure 42: P
            with the Frontier                                                               66
 Figure 43: Tfp Trends, Sector-Wide and for Traditional Crops                               67



  viii
                                                                                          ﻿Contents




Figure 44: Sources of Growth in the Traditional Crops Subsector                                   67
Figure 45: Technology Gap Ratio                                                                   75
Figure 46: Technical Efficiency Distributions: 1982, 1999, 2007                                   78
Figure 47: Changes in Farm-Level Technical Efficiency by State                                    78
Figure 48: Crop-Level Efficiency Estimates: 1999-2007                                             79
Figure 49: Regional Differences in Efficiency by Geographical and Agro-Ecological Zone            80
Figure 50: Farm Economic and Allocative Efficiency                                                80
Figure 51: Allocative Efficiency at the Crop and Farm Levels                                      81
Figure 52: Changing Farm Size and Profitability Relationship, 1982–2007                           84
Figure 53: Revenue and Costs per Acre by Farm Size                                                85
Figure 54: Cost Structure by Farm Size                                                            85
Figure 55: Growth in India’s Rice and Wheat Yields has Slowed (10-Year Trend Rates)               87
Figure 56: Global Rice and Wheat Yields have Plateaued                                            88
Figure 57: Yield Gaps, All India                                                                  89
Figure 58: Yield Gaps, Punjab                                                                     90
Figure 59: Current Yield Gaps and Progress in Reducing Yield Gap                                  90
Figure 60: Experimental Rice Yields for Released Varieties, 1990–2010                             92
Figure 61: Experimental Wheat Yields for Released Varieties, 1990–2010                            92
Figure 62: Simulated Impact of Temperatures on Grain Yields                                       93
Figure 63: Agricultural Gdp: Share of Livestock, 1950/51–20110/11 (In 2004–05 Prices)         100
Figure 64: Milk Yield Variability by State                                                    103
Figure 65: Share of Agriculture in Gross Capital Formation                                    117
Figure 66: Trends in Public Investment (2004/05 Prices)                                       119
Figure 67: Subsidies and Public Investment as a Share of Agricultural Gdp                     119
Figure 68: Annual Growth in Farm Equipment                                                    121
Figure 69: Composition of Household Assets (Other than Real Estate), Real 1999 Prices         121
            rends in Share of Irrigated Area and Yields (Left Panel) and Decadal
FIgure 70: T
           Trend Growth Rates in Share of Irrigated Area (Right Panel)                        123
Figure 71: Virtual Water Imports and Public Procurement of Rice                               124
Figure 72: Impact of Nutrient Imbalance on Land Productivity                                  125
Figure 73: Composition of Food Exports, 2012/13                                               144
Figure 74: Dualistic Structure of the Agro-Processing Sector                                  145



                                                                                              ix
Republic of India: Accelerating Agricultural Productivity Growth




 Figure 75: Share of Agro-Processing in Organized Manufacturing                                  145
 Figure 76: Share of Agro-Processing in Unorganized Manufacturing                                146
 Figure 77: State Poverty Rates and Shares in Population, Employment                             146
 Figure 78: Rural-Urban Ratios in the Unorganized Sector                                         147
 Figure 79: Trends in Labor Intensity                                                            148
 Figure 80: Ratios of Indicators in Food to Non-Food-Processing Industry                         149
 Figure 81: Tfpg in Organized and Unorganized Food Manufacturing, 2000–09 and 2000–05            154
 Figure 82: Relationship between Growth in Gva and Tfp                                           155
 Figure 83: Technical Efficiency Across States                                                   156
 Figure 84: Change in Labor and Land Productivity for Lis and Non-Lis, 1990–2008                 162
 Figure 85: Average Yields of Major Crops in Selected States (2009/10, Mt/ha)                    163
             ice Yield Gaps in Bihar and Odisha, 2009 (Difference between District
 Figure 86: R
            Average Yields and Frontline Demonstration Yields, by Variety)                       163
 Figure 87: Sources of Growth in Bihar and Odisha, 1981–2010                                     164
 Figure 88: Sources of Output Growth in Selected States, 1981–2008 (%)                           165
 Figure 89: Tfp in Traditional Crops, Bihar and Odisha, 1981–2008                                166
 Figure 90: Extent of Irrigation and Sources of Irrigation Water in Bihar and Odisha, 2008 (%)   169
Boxes
 Box 1: Structural Transformation of Economies                                                    43
 Box 2: Agricultural Research Organization, Investment, and Capacity in India                     95
         esolutions from the Bhubaneswar Declaration on Higher Agricultural
 Box 3: R
        Education in India                                                                        96
 Box 4: Atma—Steps Toward Demand-Driven, Pluralistic Extension                                    98
 Box 5: Progress in Market Reforms in Maharashtra                                                138




           Regional Vice President	         :        Philippe H. Le Houerou
           Country Director	                :        Onno Ruhl
           Sector Director	                 :        John Henry Stein
           Sector Manager	                  :        Simeon K. Ehui
           Task Team Leader	                :        Madhur Gautam




   x
                  Preface




T
      he motivation for this study is to gain a      in Eastern India, a priority area for the
      deeper understanding of the constraints        government owing to the concentration of
      on agricultural productivity growth,           Low-income States in the region; and (iii) an
the central challenge facing agriculture in          assessment of the implementation of specific
India. The study aims to inform the ongoing          programs—for example, the Rashtriya Krishi
debates on strategic issues of relevance to the      Vikas Yojana (National Agriculture Development
Government of India’s development objective of       Programme, RKVY), the Agricultural Technology
inclusive and sustainable growth.                    Management Agency (ATMA), National Food
                                                     Security Mission (NFSM), and the National
A comprehensive study to tackle all of the           Horticultural Mission—and policy reforms,
numerous issues would be an enormous task.           such as the Agricultural Produce Market
The present study is less ambitious. India           Committees (APMCs).
is a large and heterogeneous country. No
individual study is likely to be sufficiently        These initial consultations were followed by
broad and deep to do justice to the many and         a national workshop in November 2011 in
complex issues surrounding its agricultural          New Delhi, organized by the Indira Gandhi
development. For that reason, this study was         Institute of Development Research (IGIDR) and
designed with particular care to complement,         the Institute for Human Development (IHD) in
build on, and avoid redundancy with the array        collaboration with the Planning Commission
of ongoing analytical efforts and recently           and the Ministry of Agriculture. The
completed studies that address some of the           preliminary papers and invited presentations
other pressing agricultural issues, such as          provided important input on key issues in
water and irrigation.                                raising agricultural productivity. Following the
                                                     workshop, the background papers were refined,
As a result of extensive consultations, the scope    and additional detailed studies on selected
of this study was deliberately narrowed to focus     topics were commissioned. That information
on drivers of productivity, including: (i) factors   formed the basis for this report.
explaining the patterns of productivity growth;
(ii) strategies to best support resource-poor        To ensure its relevance and to take advantage of
areas and small and marginal farmers with            the large knowledge base on Indian agriculture,
specific focus on promoting faster growth            the team collaborated closely with the major




                                                                                                 xi
Republic of India: Accelerating Agricultural Productivity Growth




think-tanks and research organizations             Research (NCAER). Researchers from several
working in India, including IGIDR, the             other academic and research organizations
International Food Policy Research Institute       contributed to background papers and analyses.
(IFPRI), the National Centre for Agricultural      The Bill and Melinda Gates Foundation was also
Economics and Policy Research (NCAP), and          a partner in the study, with a specific interest in
the National Council for Applied Economic          the low-income areas of Eastern India.




  xii
                  Acknowledgments




T
      his study is an outcome of a joint effort by   Institute of Development Research (IGIDR,
      the World Bank and the United Nations          Mumbai); the Indian Council for Agricultural
      Food and Agriculture Organization (FAO),       Research; the Directorate of Rice Research
in collaboration with the Ministry of Agriculture    (Hyderabad); the Directorate of Wheat Research
and the Planning Commission, Government              (Karnal); the National Center for Agricultural
of India. Parts of the study were funded by the      Economics and Policy Research (NCAER, New
United Kingdom through the UK Department             Delhi); the FAO Regional Office for Asia and
for International Development and World Bank         the Pacific (Bangkok); the International Water
Strategic Partnership for India III, and by FAO      Management Institute (New Delhi); and the
through the Technical Cooperation Programme          World Economic Forum (Geneva).
Project, TCP/IND/3301—Policy Options
and Investment Priorities for Accelerating           The study team also gratefully acknowledges
Agricultural Growth in India. The two state          the generous contributions of Keith Fuglie
case-studies on Eastern India were financed          in making available the global Total Factor
by the Bill and Melinda Gates Foundation and         Productivity (TFP) database on agriculture; P.K.
undertaken by the International Food Policy          Joshi (IFPRI, New Delhi) for access to various
Research Institute (IFPRI, New Delhi); the Asian     databases on Indian agriculture, including the
Development Bank (New Delhi) undertook               growth decomposition database; Parthasarthy
the analysis of the pro-poor value chains.           Rao and the Village Dynamics in South Asia
The views expressed in this report do not            team at ICRISAT for providing the District
necessarily reflect official policies or views of    Level database for India; Pradhuman Kumar
the Government of the United Kingdom, FAO,           for the historical Cost of Cultivation and
the Bill and Melinda Gates Foundation, or the        research expenditures databases; Pramod
Asian Development Bank, however.                     Aggarwal for the simulations of potential rice
                                                     and wheat yields; Seema Bathla for the data
A study of this kind requires significant input      on the historical wholesale prices and public
and support in the form of analytical input,         expenditures on infrastructure; Nicholas Rada
data, and information. For that support, the         for estimates of agricultural sector TFP; Tony
study team wishes to thank IFPRI (New Delhi          Fischer and Derek Byerlee for estimates of
and Washington, DC); the International Crops         global yield gaps; and Kailash Pradhan and
Research Institute for the Semi-Arid Tropics         Sudhir Kumar for the data management and
(ICRISAT, Hyderabad); the Indira Gandhi              processing of the NCAER-REDS database.



                                                                                                  xiii
Republic of India: Accelerating Agricultural Productivity Growth




The core team initiating the study included        Chand, S. Mahendra Dev, Ashok Gulati,
Madhur Gautam, Deepak Ahluwalia, Severin           P.K. Joshi, Mukesh Khullar, Uma Lele, Prabhu
Kodderitzsch, and Shilpa Phadke (from the          Pingali, V.V. Sadamate and Abhijit Sen advanced
World Bank); Gavin Wall, Bhaskar Goswami,          the work significantly, as did the insights from
Syed Saifullah, and Sumiter Broca (from FAO);      Pramod Aggarwal, Raman Ahuja, Ganesh-
and S. Mahendra Dev (IGIDR) as lead technical      Kumar Anandam, Suresh Babu, Seema Bathla,
advisor. Madhur Gautam, Bhaskar Goswami,           Pratap Birthal, Liz Drake, Tony Fischer, Keith
S. Mahendra Dev, and Srijit Mishra coordinated     Fuglie, Nilabja Ghosh, Raj Gupta, Girish
the background studies. IGIDR and the              Kumar Jha, Anjani Kumar, Pradhuman Kumar,
Institute for Human Development (New Delhi)        Shaik Meera, Srijit Mishra, Hari Nagarajan,
organized the initial workshop in Delhi under      Tuu-Van Nguyen, Suresh Pal, Manoj Panda,
the leadership of S. Mahendra Dev, Alakh           Sumit Mazumdar, Gokul Patnaik, Nicholas
Sharma, Srijit Mishra, and Sumit Mazumdar.         Rada, Anil Rajput, Shobha Rani, Baskar Reddy,
Madhur Gautam wrote the final report with          Mangal Sain, C.S.C. Sekhar, Alakh Sharma,
substantive input from Deepak Ahluwalia,           Arun Kumar Sharma, Indu Sharma, R.K.
Syud Amer Ahmed, Varun Kshirsagar, and             Sharma, Jit Shrivastava, R.B. Singh, Randhir
Helen Leitch, under the overall guidance           Singh, M. Srinivas Rao, Sukhadeo Thorat,
of Simeon Ehui, Sector Manager for                 V.C. Viraktamath, and J.S. Yadav. Finally, the
Agriculture, Irrigation, and Natural Resources,    study team appreciates the suggestions and
South Asia Region and Onno Ruhl, Country           support of colleagues at the World Bank Group
Director for India.                                and FAO, including Luis Alberto Andres, Ulrich
                                                   Bartsch, Deepak Bhattasali, Dan Biller, Edward
At various stages, this study benefited from       Cook, Paramita Dasgupta, Klaus Deninger,
the perspectives and other contributions of        Grahame Dixie, Anju Gaur, Ejaz Ghani, Neeraj
a large number of people. Foremost among           Gupta, Kalpana Kocchar, William Magrath, Ajay
them are the many authors of the technical         Markanday, William Martin, Elliot Mghenyi,
background papers and notes listed in the          Grant Milne, Rinku Murgai, Keith Oblitas,
first part of the references. The feedback and     Manivannan Pathy, Jeeva Perumalpillai-
comments from the many participants at the         Essex, Shilpa Phadke, Giovanna Prennushi,
Delhi workshop was helpful in refining the         Martin Rama, Ranjan Samantaray, Animesh
background papers and also helped shape the        Shrivastava, Paul Sidhu, Dina Umali-Deininger,
agenda for the subsequent analysis used in         Harsh Vivek, Vinay Vutukuru, Melissa Williams,
this report. Comments and suggestions from         Winston Yu, Roberto Zagha, (all World Bank
peer reviewers Derek Byerlee, Dina-Umali           Group) and Peter Kenmore, Vinod Ahuja and
Deininger, Peter Hazell and Robert Townsend        Sumiter Broca (FAO). Sarita Rana and Lilac
improved the report. Advice and discussions        Thomas provided excellent administrative
with S. Ayyappan, Shawki Barghouti, Ramesh         support. Kelly Cassaday edited the report.




  xiv
                  Acronyms and Abbreviations




AI        Artificial Insemination            ECA       Essential Commodities Act
APC       Agricultural Policy Costs          ENSO      El Niño-Southern Oscillation
APEDA     A
           gricultural and Processed Food   F&V       Fruits and vegetables
          Products Export Development        FAO       F
                                                        ood and Agriculture Organization
          Authority                                    of the United Nations
APMC       gricultural Produce Market
          A                                  FDI       Foreign direct investment
          Committee
                                             FMD       Foot and mouth disease
           gricultural Produce Markets
APM (D&R) A
                                             FY        Farm yield
          (Development and Regulation) Act
                                             g         Gram
AS        Assam
                                             GA        Growth accounting
ASI       Annual Survey of Industries
                                             GCI       Global Competitiveness Index
ATMA       gricultural Technology
          A
          Management Agency                  GDP       Gross domestic product

AY        Attainable Yield                   GJ        Gujarat
                                             GOI       Government of India
BH        Bihar
                                             GSDP      Gross state domestic product
BIIC      Brazil, India, Indonesia, China
                                             GVA       Gross value added
Bt        Bacillus thuringiensis
                                             GWh       Gigawatt hour
CF        Contract Farming
                                             ha        Hectare
CoC       Cost of Cultivation (database)
                                             HP        Himachal Pradesh
CRS       Constant returns to scale
                                             HY        Haryana
CSO       Central Statistical Organization
                                             IBTC      Input-biased technical change
DEA       Data Envelopment Analysis
                                             ICAR      I
                                                        ndian Council of Agricultural
DME       D
           irectory Manufacturing
                                                       Research
          Establishment
                                             ICRISAT   I
                                                        nternational Crops Research
DWR       Directorate of Wheat Research
                                                       Institute for the Semi-Arid Tropics
DRR       Directorate of Rice Research
                                             ICT       I
                                                        nformation and communication
EC        Efficiency change                            technology


                                                                                         xv
Republic of India: Accelerating Agricultural Productivity Growth




IFPRI        nternational Food Policy Research
            I                                       NSS            National Sample Survey
            Institute                               OAME           O
                                                                    wn Account Manufacturing
IGIDR       I
             ndira Gandhi Institute of                            Enterprise
            Development Research                    OBC            Other Backward Castes
IHD         Institute for Human Development         OBTC           Output-biased technical change
IMD         India Meteorological Department         OR             Odisha
JH          Jharkhand                               PB             Punjab
KE          Kerala                                  PEC            Pure technical efficiency
kg          Kilogram                                PPP            Public-private partnership
km          Kilometer                               PY             Potential yield
KT          Karantaka                               q              Quintal
KVC          rishi Vikas Cooperative
            K                                       QSR            Quick service restaurant
            (Sholapur, Maharashtra)
                                                    RBH            Rural business hubs
KVK         Krishi Vigyan Kendra
                                                    RDI            Relative Diversity Index
LIS         Low-income States
                                                    RKVY            ashtriya Krishi Vikas Yojana
                                                                   R
LQ          Location Quotient                                      (National Agriculture Development
MH          Maharashtra                                            Programme)
MNREGA       ahatma Gandhi National Rural
            M                                       RJ             Rajasthan
            Employment Guarantee Act                RRA            Relative Rate of Assistance
MP          Madhya Pradesh                          Rs             Rupees
MSP         Minimum Support Price                   RY             Realizable yield
MT          Metric ton                              SC/ST          Scheduled castes/scheduled tribes
MTC         Magnitude or neutral component          SEC            Scale efficiency
NCAER        ational Council for Applied
            N                                       TC             Technical change
            Economic Research
                                                    TE             Triennial Ending
NCAP         ational Centre for Agricultural
            N
                                                    TFP            Total factor productivity
            Economics and Policy Research
                                                    TFPG           Total factor productivity growth
NDME         on-directory Manufacturing
            N
            Establishment                           TN             Tamil Nadu
NFSM        National Food Security Mission          UK             Uttarakhand
NGO         Nongovernmental Organization            UP             Uttar Pradesh
NIC         National Industries Classification      WB             West Bengal
NPK         Nitrogen, phosphorus, potassium         WPI            Wholesale Price Index
NRA         Nominal Rate of Assistance




  xvi
                   Accelerating Agricultural Productivity
                   Growth in India
                   OVERVIEW




I
    n the past 50 years, Indian agriculture          to be ambitious goals. In the past, India’s
    has undergone a major transformation,            unwavering focus on food production helped
    from dependence on food aid to becoming          it to achieve self-sufficiency, but a legacy of
a consistent net food exporter. The gradual          that effort is a complex web of policies and
reforms in the agricultural sector (following        institutions that now arguably constrain more
the broader macro-reforms of the early 1990s)        robust, sustainable agricultural growth, limit
spurred some unprecedented innovations               the performance of agricultural markets, and
and changes in the food sector driven                discourage much-needed diversification. The
by private investment. These impressive              natural resources that support productive
achievements must now be viewed in light of          agriculture (namely land and water) are
the policy and investment imperatives that           declining in quality, and competition for
lie ahead. Agricultural growth has improved          them is intensifying. Rainfall remains a major
in recent years (averaging about 3.5 percent         source of volatility in Indian agriculture. Heavy
since 2004/05), but at a long-term trend             public investments, particularly in irrigation
rate of growth of 3 percent, agriculture has         and technology, have helped to offset the
underperformed relative to its potential.            worst effects of weather, but the deceleration
The pockets of post-reform dynamism that             of growth in the late 1990s and early 2000s,
have emerged evidently have not reached              persistent increases in food prices in recent
a sufficiently large scale to influence the          years, and declining water tables have revived
sector’s performance. For the vast population        concerns over food security. Climate change will
that still derives a living directly or indirectly   almost certainly magnify the challenges and
from agriculture, achieving “faster, more            expectations for agriculture.
inclusive, and sustainable growth”—the
objectives at the heart of the Twelfth Five          India is a large, heterogeneous country. One
Year Plan—depends critically on simultaneous         study alone cannot address the multitude of
efforts to improve agriculture’s performance         issues surrounding agriculture, certainly not
and develop new sources of employment for            in sufficient depth to be meaningful. This
the disproportionately large share of the labor      study was designed through a broad-based
force still on the farm.                             consensus to focus on strategic issues related
                                                     to agricultural productivity. It gives particular
Maintaining India’s hard-won food security           attention to the dynamics and sources of
and achieving shared prosperity are proving          productivity growth, the sustainability of



                                                                                                   xvii
Republic of India: Accelerating Agricultural Productivity Growth




growth, and areas where the potential for             natural resource base, placing the hard-
growth has been overlooked, all with a view to        won productivity gains of the past—and the
informing the debates on strategic priorities         future—at risk. Other parts of the country face a
and policy interventions.                             dilemma of another sort. Although they are far
                                                      less agriculturally developed, these Low-income
The scope of this study is broad in the sense         States (LIS) could unleash considerable growth
that it marshals considerable empirical evidence      in agricultural productivity, yet weak public
and analyses to address those issues. Yet the         investments, undeveloped markets and weak
scope is restricted in the sense that the study       institutional and governance capacity have long
does not address all of the issues. A wealth of       stood in the way.
knowledge exists (and continuing analytical
work proceeds) on other major strategic               A major puzzle seems to be the co-existence of
issues—water and irrigation management, food          widespread undernourishment and rising food
grain management, and public expenditures             prices on the one hand, and record production
on agriculture, for example—and the findings          levels of food and overflowing stocks on
of this study must be seen in that context. The       the other. Across India, diversification into
lack of sufficient quality data, and often the lack   higher-value crops and livestock products has
of access to such data, also prevent some issues      proceeded too slowly to increase agricultural
from being explored in greater depth. Finally,        growth appreciably. The supply of high-value
some important issues require more focused            commodities has not kept pace with demand
and dedicated analysis, such as food safety           generated by rising incomes and urbanization,
and quality standards, agricultural trade, and        resulting in rapid increases in their prices.
food price increases. This relationship between       Cereal prices have again started to rise, adding
longer-term strategic issues and contemporary         to an agricultural conundrum in which per
concerns, such as water resource management           capita availability of food is falling even as per
and food prices, are highlighted in this study        capita production of cereals is at all-time highs
through the prism of productivity, but they           and rising, and domestic markets are effectively
too require further analysis to fully address the     insulated from global market pressures. Rising
underlying issues.                                    Minimum Support Prices (MSPs) create a cereal
                                                      supply response, but the increased production
                                                      is diverted to stock silos rather than benefiting
The Conundrums of Contemporary
                                                      consumers through lower prices. Thus MSPs
Agriculture                                           drive domestic cereal prices—especially
Contemporary Indian agriculture presents              producer prices—higher, while a combination
a number of seeming contradictions and                of trade and storage policies stabilize them.
conundrums. India’s traditional breadbaskets          The resulting low risk-to-return ratio for cereals
face the food security–sustainability tradeoff.       creates strong incentives for farmers to produce
The irrigated rice- and wheat-producing areas         cereals rather than other (more risky) crops,
appear to be facing diminishing returns to the        limiting diversification and income growth,
technology that sparked the green revolution.         putting further pressure on prices—this
These areas are singularly focused on increasing      time through the higher-valued non-cereal
production, often at the cost of mining the           commodities.



 xviii
                                                                                             ﻿OVERVIEW




The short-term welfare impacts appear to be          is less obvious. Because the slowdown coincided
contained, as rural wages have risen rapidly         with the agricultural policy reforms of the
in recent years, compensating for rising             mid-1990s (following the general economic
prices. Rising real prices benefit the net-selling   reforms of the early 1990s), a natural conclusion
farmers, and rising real wages are good for          often drawn is that the reforms triggered the
workers, but both trends have implications           weakness. The influence of rainfall on Indian
for the sustainability of the ensuing growth.        agriculture is widely acknowledged—60 percent
The critical question is whether the changed         of it remains rainfed—but rainfall’s role in
incentives afforded by rising real output prices     patterns of growth has not been fully explored.
are accompanied by growth in productivity—
are the sources of that growth sustainable?          At the district level, a credible association
                                                     emerges between rainfall shocks and
                                                     productivity (defined as value of output
Trends in Agricultural Performance                   per hectare). Aggregated up to the national
at the National and Subnational                      level, these data show a strong association of
Level                                                an unusual sequence of sustained negative
                                                     rainfall shocks and the prolonged stagnation
Over the past six decades, agriculture grew          in productivity growth between 1999 and
at a steady but modest 3.0 percent, changing         2004. The weather-induced weakness provides
imperceptibly but becoming relatively more           an explanation for the deviation in trends for
stable. After almost two decades of sustained        several key indicators during the early 2000s,
expansion (with growth peaking in the early          including the slowdown in wages identified in
1990s at about 3.6 percent), growth decelerated      previous studies. By extension, poor rainfall in
from 1996/97 to 2004/05. This prolonged              the more recent years may partly account for
slowdown was widespread, with few exceptions.        the sluggish supply response in noncereal crops,
The most recent period—from 2004/05                  which are found to be more susceptible to such
onwards—shows a marked, equally widespread           shocks, and perhaps partly explain the upward
return to growth of 3.5 percent per year.            pressure on their prices.

Performance and rainfall. What started               The sensitivity of crop productivity to rainfall
agriculture on its long decline after 1997, and      varies considerably across districts, reflecting
what spurred the equally decisive turnaround         the availability of irrigation to compensate
after 2004? Explanations for the slowdown            for annual rainfall deficits. Over the long run,
include slow generation of new technologies,         districts in the Semi-Arid Temperate Zone are
poor dissemination of existing ones, weak            generally less vulnerable to rainfall shocks
and inefficient institutions, poor governance,       and relatively more productive on average. But
and inadequate investment in public goods.           the sustained rainfall shocks between 1997
Irrigation, terms of trade, and technology are       and 2004 hit those districts the hardest, with
the major determinants of agricultural GDP           a large aggregate and cumulative impact on
growth. These factors are crucial for long-term      agricultural productivity. The cyclical as well
growth, but their immediate role in the post-        as random anomalies that are characteristic of
reform slowdown and the subsequent recovery          rainfall in the Indian sub-continent call for both



                                                                                                  xix
Republic of India: Accelerating Agricultural Productivity Growth




ex ante risk mitigation and ex post adaptation/         near to medium term, and the anticipated
management strategies.                                  intensification of the spatial and inter-annual
                                                        variability of rainfall over the very long term.
Coming back to the question of role of policy
reforms in agriculture’s performance, a key             Dynamics of agricultural productivity
question is what growth rates might have                growth. Wide differences in performance
prevailed had rainfall been normal from the             across states in the 1980s appear to have
1960s to 2010. A simple simulation reveals a            disappeared since 2004/05, suggesting that
sharp deviation between the actual and the              growth may now be more inclusive, with
counterfactual (simulated assuming normal               lagging states starting to grow at par with other
rainfall, all else remaining the same) trends           states. But within states (and agro-ecological
starting in the mid-1990s, suggesting that had          zones), performance has varied widely across
rainfall been normal, the growth trajectory             districts over time. A few districts have done
would have been significantly different and             well but most have not, and the relative
may possibly have been higher than the actual           rankings across districts are mostly preserved
observed trend, potentially ushering in a               with persistent large differences. The lagging
much-desired, positive structural change. An            districts are not growing fast enough yet to
important implication of these findings is that         achieve real convergence. For example, in
there is little evidence to suggest that the policy     2007–08, productivity was 50 times higher in
reforms of the mid 1990s had a significant              the most productive versus the least productive
adverse impact on agriculture, as may be                districts. Within LIS, increased growth in some
tempting to infer from the observed growth              districts widened inequality—the exception
trend. Another, perhaps more important,                 being Odisha—but in the more advanced
implication is the urgency of improving                 agricultural states, lagging districts are catching
agriculture’s resilience to the shifting trends         up with the others, indicating convergence in
in anomalous weather patterns over the                  growth rates.

Messages and Implications

There is little evidence to support the assertion that the policy reforms of the 1990s likely had a
negative impact on agriculture. At the same time, the long-term strategic public investments in
technology and irrigation have been very successful in increasing food production and mitigating
the worst effects of rainfall shocks. The large aggregate and cumulative impact of a series of negative
rainfall shocks between 1999-2004 serves a very important purpose—as a harbinger of the potential impacts
of changing weather patterns over the short to medium term and the highly uncertain climate change
outcomes over the long term. The experience underscores the urgency to address the current strategic
issues. It highlights the critical importance of more efficient water management practices to weatherproof
agriculture; develop strategies and make investments to mitigate climatic variability and increase resilience
amid climate change (climate-smart technologies, sustainable irrigation, water harvesting and watershed
development); improve markets and marketing to allow real-time risk sharing across states and districts in
response to emerging market signals; and diversify and stabilize sources of income (outside the crop sector)
through livestock and productive nonfarm employment. It is essential to quantify the costs and benefits of
alternative ex ante risk-mitigation strategies and ex post risk-sharing strategies—issues for further
detailed study.




  xx
                                                                                                      ﻿OVERVIEW




To more closely examine performance and                  low-yield/growth districts, unlike the low-yield/
drivers of productivity, districts are classified by     stable districts, switched area from low-value
growth typology based on their initial (1970s)           crops to higher-value crops such as cotton,
yields (high or low) and their subsequent                horticultural crops, and soybeans.
performance (relatively stable or fast-growing)
over the study period (1970–2008). Most                  District-level indicators to explain changes
growth in agricultural productivity occurred in          in agricultural productivity are limited, but
northern and southern districts. Agro-ecological         the available data convey some important
conditions and rainfall anomalies by themselves          information. Markets per capita declined in
do not sufficiently explain the disparities              low-yield districts relative to the others, likely
among districts. What sets the growth districts          reducing producer incentives and productivity
apart from the others is better access to                growth. Road density has improved in the
irrigation (and fertilizer use, which is influenced      laggard districts, bringing them on par with
by irrigation). In addition to intensification,          the other districts. But while roads continue
diversification into nontraditional and higher-          to show significant positive impact in the
value crops is also a major driver of growth             high-yield areas, they have not yet translated
in the high performers. For example, the                 into improved productivity in the lagging


Messages and Implications

At the subnational level, growth has been pervasive—but not sufficiently broad-based. While the
pace of agricultural growth appears to be converging somewhat across states, at the district level the
expansionary phase of the 1980s and early 1990s was not very inclusive. A few lagging districts “caught
up” with the better-performing districts, but a substantial number fell further behind, despite registering
positive but low levels of growth.
Differences across districts are rooted in strategy and the enabling environment—and that is where
policy attention is needed. Weather alone does not explain the differences in performance as weather
shocks affected the productive districts disproportionately. At the same time, the findings caution against
a “silver bullet” approach, highlighting the complementarity of policies and investments tailored to
particular circumstances. For example, roads and improved seeds alone do not account for differential
performance among districts.
The main drivers of productivity at the district level—and hence the key entry points for action—
appear to be markets and irrigation. Market density has fallen more in the low-yield/stable districts
than in the others, likely constraining productivity growth (through producer incentives). Irrigation and
the associated adoption of fertilizer have contributed to significant changes in productivity in the growth
districts. Improved seeds, the other key element of the green revolution technology, have spread faster
and wider, but by themselves they have not narrowed the productivity differentials across districts. The
rapid expansion in irrigation occurred mainly through groundwater extraction, well-known among policy
instruments as a double-edged sword. The recurrent theme of sustainable water management emerges
as a policy priority, with the important lesson from the faster-growing districts, consistently appearing
across the low–high yield typologies, that diversification needs to be prioritized for a possible win-win
strategy. That said, some areas will inevitably have limited prospects for irrigation, and their agro-ecological
endowments may limit the scope for certain types of agriculture. Localized strategies will be needed for
these areas to identify viable opportunities, including livelihood options outside of agriculture.




                                                                                                           xxi
Republic of India: Accelerating Agricultural Productivity Growth




districts, calling attention to the importance         nonagricultural workers. The widening of
of complementary investments. A slightly               this gap is a worry for policy makers. The low
higher share of the population in low-yield            productivity of a large proportion of the labor
districts was rural to begin with; this indicator      force places a heavy tax on overall well-being
of urbanization has changed relatively slowly in       and shared prosperity. But how atypical is
the districts where yields were initially low but      India’s experience?
shows no obvious links with the pace of growth
within the low- or high-yield cohorts. Similarly,      Evidence suggests that developing countries
literacy remains marginally lower in the low-          are now taking longer to reach their “turning
yield districts, but this basic measure of human       point”—the point at which the inter-sectoral
capital development has also improved more             labor productivities start to converge. The
rapidly in those districts, again without a clear      implication is that the development context
association with productivity growth.                  is changing, and it is increasingly harder
                                                       to absorb labor outside agriculture. India
                                                       seems to be experiencing this phenomenon
An International Perspective on
                                                       and is behaving no differently than the
India’s Structural Transformation                      average developing country (in a cohort of
The slow pace of India’s structural                    88 developing countries). Consequently, India
transformation (that is, the decline in the            must pay particular attention to accelerating
share of labor in agriculture relative to              the pace of labor absorption outside of
the decline in the share of agriculture in             agriculture, and it must redouble efforts to
aggregate GDP) is reflected in the large gap           increase labor productivity within agriculture.
in productivity between agricultural and               Making agricultural labor more productive is

Messages and Implications

Technology has been a consistent driver of productivity in all four countries. Investing in technology
is a globally proven and tested strategy for promoting productivity growth. Brazil, India, and China are
among the largest investors in public research, and all have benefited significantly from such investment.
China’s and Brazil’s investments, however, have been relatively more effective, making them global leaders
in agricultural innovation. In Indonesia, the benefits of technology have also been pivotal, but they have
accrued through open trade rather than domestic research.
A second major driver was agricultural diversification, both for domestic and export markets, supported
by appropriate technology, policy, institutional reforms, and investments. Brazil, Indonesia, and China all
benefited significantly more than India from openness in trade.
The backbone of China’s and Brazil’s more rapid transformation has been a more predictable
enabling and policy environment. With a strong record of implementation, continuous innovation in
public sector management for agriculture and rural development, and more effective decentralization in
decision making, they have achieved significantly greater productivity growth. For example, major and
fundamental reforms have greatly increased water-use efficiency in Chinese agriculture. Conducive policies
paid significant dividends in production efficiencies and diversification, and better access to technology,
whether from the international public research system, national research organizations, or the private
sector (notably in relation to genetically modified crops).




 xxii
                                                                                                   ﻿OVERVIEW




imperative in any case, because a declining             A Micro-level Perspective on India’s
farm population will have to meet the                   Structural Transformation
consumption and raw material requirements of
a growing nonfarm population.                           A unique longitudinal study at the village
                                                        level in Bihar, one of India’s poorest states,
In this context, comparisons among four large           draws attention to major changes at the micro
developing countries, Brazil, Indonesia, India,         level. Semi-feudal production relations—long
and China, are useful. Starting with comparable         associated with Bihar’s agriculture—have
conditions in 1961, China and Indonesia                 virtually disappeared, while numbers of poor
achieved more success in reducing poverty               peasants, shares of casual landless agricultural
and improving rural well-being. Brazil was              labor, and the proportion of nonagricultural
relatively more advanced from the start and             households all increased. Nonagricultural
has continued to perform well. While Brazil             income, dominated by remittances, is the main
has reached and China has almost reached                source of income for all economic classes.
their turning points, structural transformation         Poor peasants have similar sources of income,
in India and Indonesia has been slower. If the          but they largely undertake nonagricultural
status quo prevails, their turning points are           production activities. The middle and big
projected to be at least two decades away.              peasants have a higher share of agricultural
                                                        income, but it is still less than half their total
China is a more relevant comparator in terms of         income.
the scale of farming. The main difference with
India is that despite having a larger share of its      Another aspect of structural transformation
workforce in agriculture, China has seen much           is the significant diversification in men’s
higher growth in agricultural value-added, with         occupations. Levels of migration out of the
significantly more rapid technological change           village are high, and migration income is
and diversification, and a much greater reliance        the largest share of household income. At
on efficiency than input use as the main driver         least initially, migration seems to have been
of growth. With more rapid increases in labor           a response to the lack of opportunity in
productivity, living standards among China’s            local labor markets. Migration affects rural
agricultural population improved much faster.           production systems by pushing up local wages,

Messages and Implications

Even villages in poor and backward areas, typical of Bihar, are experiencing significant agricultural
and socioeconomic change. Nonagricultural incomes, particularly remittances, dominate household
incomes. But agriculture, with a steady increase in yields (faster than the national average for rice and
wheat), has not been stagnant. Real incomes appear to be rising, along with agricultural productivity, and
agricultural diversification is yielding significant benefits. Structural transformation is more challenging
in Bihar than elsewhere in India. Despite high levels of migration (Bihar is unusual in this respect), a
substantial proportion of the population remains in agriculture; consequently, the person-land ratio is
significantly worse, keeping the productivity of agricultural labor low. Concerns about the sustainability of
migration as the major engine of continued growth call for more rapid nonfarm and on-farm diversification
as sources of income growth.




                                                                                                        xxiii
Republic of India: Accelerating Agricultural Productivity Growth




promoting labor-saving cultivation techniques,          the recovery (after 2003), but since 2007, area
and increasing the feminization of agricultural         expansion has slowed as expected. Importantly,
labor markets.                                          yields’ contribution to growth diminished
                                                        considerably, and diversification remains
                                                        modest despite rapidly changing diets and
Did the Drivers of Growth Change                        rising commodity prices.
Qualitatively in the 2000s?
To assess whether the recent recovery in                Evolution of Productivity at the
agricultural growth can be sustained, it is
                                                        National, State, Household, and
vital to learn whether the drivers of growth
changed qualitatively in the mid-2000s.                 Farm Levels
Agricultural growth is increasingly driven by           Given India’s binding land constraint,
the rising shares of high-value commodities             agricultural growth depends on making
in value terms, but food grains still occupy            land (for crops) and animals (for livestock
two-thirds of the cropped area, and the shares          products) more productive. In the case of land,
of rice and wheat are unchanged. Except                 productivity, often equated to yields, can be
for cotton, yields of high-value crops have             enhanced through intensification (using more
not increased significantly, raising concerns           inputs per hectare), through technological
about the sustainability of their growth. A             advances (better inputs), and/or improved
decomposition of growth confirms these                  efficiency (using inputs more effectively).
apprehensions.                                          Total factor productivity (TFP) captures the
                                                        contributions of technology and efficiency and
Yields dominated growth until the mid-                  provides a summary measure of the health
1990s, as green revolution technology spread.           of the production system. For growth to be
Diversification has been a consistent but               ecologically and economically sustainable, TFP
moderate contributor to growth. Since the               must improve.
1980s, diversification consistently accounted
for about one-quarter of growth, somewhat               To build a convincing body of evidence, TFP
less than might be expected from a rapidly              is assessed using multiple sources of data, at
transforming agriculture. Prices contributed            different levels of aggregation, and employing
increasingly to growth in the 1990s, and in             different methodologies. The various analyses
recent years they have again become the main            consistently demonstrate that productive
driver. Area and yields rebounded early in              resources are being used highly inefficiently.


Messages and Implications

Two worrying trends emerge from the analysis: The contribution of yields to productivity is declining,
and prices have emerged as the main driver of growth toward the end of the 2000s. In 2010, 55 percent
of the increase in the real value of output resulted from price increases. This finding raises concerns about
the sustainability of the recent growth spurt: Farmers (specifically the net-sellers) gain from higher prices,
but without underlying improvements in productivity, the current growth may be short-lived.




 xxiv
                                                                                                       ﻿OVERVIEW




At the national level. Analyses based on                  sustained growth even during the slowdown
different methodologies and data sources                  period). In contrast, efficiency has stagnated
suggest that the recovery in recent years has             over the long run. It improved in the 1980s but
been robust, with TFP growing at its fastest rate.        has started to decline in recent years, indicating
Previous studies noted decelerating growth after          that the gap between actual production and
the mid-1990s, but a longer time horizon (and             the realizable potential (production frontier) is
hindsight) demonstrate, as discussed earlier,             widening.
that this unusually long slowdown coincided
with an anomalous rainfall pattern. It also               Another important finding is strong divergence
means that the recent growth in TFP needs                 in TFP growth for the traditional crops (cereals,
to be interpreted with caution: the period of             pulses, oilseeds, cotton, and sugarcane) from
analysis is short, and the recent growth partially        that of the agricultural sector—which includes
reflects a rebound from a sharp decline in the            higher-valued horticulture and livestock
previous period (reaching the lowest point in             subsectors, both of which are important for
2003). It will be important to track performance          inclusive growth, and both of which have
with data from additional years to ascertain if           been major drivers of agriculture value-
the growth is robust.                                     added in recent years. Yet neither subsector
                                                          has commanded the attention of policy and
Irrigation, an important long-term driver of              public expenditure to the same extent as the
growth, appears to be contributing less to                traditional crops. Given the magnitude of the
output growth in the 2000s, perhaps reflecting            resources tied up in traditional crops, their low
limits on expansion. Increased inputs have                TFP growth (0.28 percent) contrasts starkly with
historically contributed the most to output               the sector-wide estimate (1.77 percent).
growth, but in the 2000s, TFP has been the main
source. Other emerging trends are a rise in               The continued reliance on subsidized inputs as
labor productivity and capital deepening in the           the main driver of growth in traditional crops,
most recent period.                                       with declining efficiency, is reason for serious
                                                          concern. Intensification in the lagging states
When TFP growth is decomposed into the                    where input use has been lower may be less
contributions of technology and efficiency,               worrisome, but the continued or accelerated
the key finding is that technical change                  use of inputs (often imbalanced applications of
has consistently been the primary driver of               nutrients and overexploitation of groundwater)
productivity growth over the past three decades,          in the advanced states demands corrective
growing fastest during the past decade (showing           action.


Messages and Implications

For traditional crops, which receive most of the resources devoted to agriculture, TFP growth
has stagnated. Technical change has been the primary driver of productivity growth, while efficiency
has stagnated and appears to be declining in recent years. The fact that three-quarters of the growth is
still driven by inputs for the bulk of the sector (traditional crops) raises concerns about the quality and
sustainability of agricultural growth.




                                                                                                              xxv
Republic of India: Accelerating Agricultural Productivity Growth




At the state level. TFP varies widely across             diversification (even within the more restricted
states, across crops, as well as across states for       group of traditional crops) and technology—
the same crops. The biggest gains are associated         primarily agricultural research—are the main
with new technologies for cotton and maize,              drivers of productivity growth. Contrary to
but the impact of even these new technologies            general perceptions, the analysis shows that
is not uniform across states. Among staples, rice        the contribution of research has not diminished
TFP has improved in recent years for several             over time.
states, but wheat TFP has slowed considerably,
with the exception of reform-driven Gujarat.             In contrast, the impact of agricultural extension
Sugarcane and cotton offer a telling comparison.         is considerably less visible. Persistent yield gaps,
Sugarcane experienced consistent declines in             even for rice and wheat, suggest that extension
TFP growth across all major producing states,            has not enabled producers to benefit from
while cotton experienced consistent gains.               current technology—even though traditional
The two differ in that there is substantial              crops have long been a priority of public
government intervention and significant                  extension services. This result strongly suggests
subsidies for sugarcane, whereas the cotton              that the most immediate action to enhance
sector has rapidly transformed since the private         productivity, and to counter “technology
sector introduced Bt cotton technology in 2002.          fatigue,” is to increase the effectiveness of
                                                         extension services.
What explains the wide variation in TFP across
states? As a residual measure, TFP subsumes              A telling result is that rural electrification has
many unobserved factors, making it difficult             a large and highly significant negative impact,
to assess the role of important policy levers.           strongly suggesting that the large subsidies
An analysis of the determinants of TFP, using            for electricity use in agriculture are adversely
state-level estimates and controlling for some           affecting TFP, probably by contributing to
of the confounding factors, gives results with           declining water tables (by promoting the
significant policy implications.                         unsustainable extraction of groundwater).
                                                         Similarly, the negative and statistically
The analysis confirms the importance of state-           significant impact of nutrient mix on TFP
specific factors (broadly reflecting the policy,         strongly supports the contention that subsidies
institutional, and governance environment)               promote indiscriminate applications of nitrogen
for productivity growth. Beyond these,                   that harm soils.

Messages and Implications

At the state level, policies and institutions strongly influence productivity growth, along with
diversification and technology. TFP is negatively affected by excessive or imbalanced input use,
arguably driven by input subsidies. Ironically, instead of the intended impact of boosting productivity,
subsidies may now be having quite the opposite effect, jeopardizing future productivity prospects. These
results have major implications for the environment, productivity, and sustainable growth. An immediate
area for action is more effective extension services. Extension services appear to have made little impact
on productivity, even in traditional crops, but if they can help farmers to close the existing yield gaps, they
could bring about substantial productivity gains.




 xxvi
                                                                                                ﻿OVERVIEW




At the household level. Using a completely            highlighting the roles of extension services and
different data source, the household-level            the policy environment in shaping farmers’
analysis corroborates the major emerging              choices. Smaller producers are more efficient,
finding that productive efficiency is low. It         with higher allocative efficiency, whereas larger
also shows that this inefficiency entails very        producers showed higher technical efficiency.
high costs in terms of forgone farm income.
Depending on the agro-ecological region, the          Changes in farm-level technical efficiency
average farmer is about 30–90 percent less            vary significantly by state. Farmers in Bihar
efficient than the “best in class”—those farmers      represent the case of the “poor but efficient”
who are her/his peers within the study sample.        producer—they operate at low input-output
                                                      levels, but their input use is efficient (Schultz
The micro-data confirm findings from the              1953). Punjabi farmers are equally efficient,
secondary data that technical change, rather          but they represent the high input-output case.
than economic efficiency, has played the major        The absence of any obvious agro-ecological or
role in productivity growth. Technical change         geographical correlation with performance
has generally been good in recent years, with         suggests that policies and the enabling
large gains in the formerly lagging semi-arid         environment at the state level play a strong role
topics. Average efficiency fell significantly in      in determining efficiency outcomes.
the humid and arid zones, indicating that the
average farmer is unable to keep up with the          In the most widely grown crops (rice and
fast moving technological frontier, and is falling    wheat), efficiency is lessening, and it is
farther away from it. Arid areas have been the        declining significantly for pulses and oilseeds.
most vibrant, catching up with the meta-frontier      Only “other crops”—an aggregate of crops
at a rapid rate and closing their historical          other than the cereals, pulses, and oilseeds—
technology gap with the rest of the country.          show marginal improvement in technical
                                                      efficiency. Across households there is significant
The cost of inefficiency in terms of farmers’         dispersion in efficiency levels for all crops.
net returns or profits is high. A staggering
68 percent of potential short-run profits             Finally, an important finding is the difference
(on average) are lost relative to the optimal         in economic efficiency at the crop and whole-
economic profit that was feasible in 2007 (albeit     farm level. Higher allocative efficiency at the
a slight improvement over the 73 percent lost         individual crop level suggests households are
in 1999). These losses can be attributed equally      using resources (inputs) reasonably efficiently
to technical and allocative inefficiencies,           given the relative prices they face. But when

Messages and Implications

Household-level analysis finds that inefficiency is high and very costly in terms of farm income.
Although much of the debate on Indian agricultural productivity has focused on technical or technical
production issues, these findings suggest significant scope to improve returns through more economically
rational choices. The role of policies, institutions, and the enabling environment, which is central
to micro-level economic choices, needs be brought to the front and center of the debate if India’s
farmers are to move to a more profitable agriculture.




                                                                                                    xxvii
Republic of India: Accelerating Agricultural Productivity Growth




it comes to the whole-farm level, allocative           of farmers suggests that greater access to
efficiency plunges, indicating that the bulk           productive land may be beneficial in terms
of economic inefficiency stems from farmers’           of overall agricultural productivity and more
crop choices. Household decisions to allocate          efficient resource use.
resources are determined by the relative prices
of crops and their associated risk levels, both        Among public investments, the importance of
of which are influenced considerably by the            access to pucca (paved) roads, technology, and
policy and institutional (especially market)           extension services is reaffirmed by empirical
environment.                                           results. An important insight on governance
                                                       is the impact of women’s participation in
What is driving inefficiency at the farm               Gram Sabha meetings, which raises productive
level? Among the range of factors explaining           efficiency.
productivity and efficiency, the empirical
analysis confirms the inverse land-productivity        Are smaller farms still efficient and viable?
relationship, but smaller farms are less               The 25 years of household panel data from
technically efficient. Another important               NCAER surveys provide a rare insight into the
finding is that at the margin family labor is less     relationships between farm size (land owned
productive relative to hired labor, suggesting         by households) and three levels of farming
overuse of family labor (possibly because too          returns: gross value of output per hectare,
many family members remain on the very small           gross-margins per hectare (revenues less paid
farms that are now more prevalent). Finally, as        out costs, so in essence returns to family labor
farm sizes decline through subdivision, land           and land), and profits (revenues less paid
fragmentation is becoming a problem, with              costs and imputed costs of family labor—in
direct consequences for productivity through           other words, the returns to land only). The
lower efficiency.                                      data show a significant and dramatic shift in
                                                       the relationship between farm size and net
Younger and more educated households                   revenues or profits per acre—from a strong
generally are more efficient at farming. The           inverse relationship in 1982 to an increasingly
higher productivity of the newer generation            positive relationship in 1999 and 2007.

Messages and Implications

Reforming the policy and regulatory framework governing land lease and rental markets is a high
priority to sustain productivity and growth in farm incomes. The empirical findings reaffirm the well-
established importance of factors like roads, technology, and irrigation while providing new insights. The
findings on land fragmentation and the association of profitability with farm size ultimately suggest that
some small farms may be getting too small to remain efficient or viable, despite the technical relationship
between farm size and yield. Together with the findings that perhaps too much family labor remains and is
used on small (and ever smaller) farms, these findings lend new urgency to reforming the tenancy laws and
legalizing land lease markets. More efficient lease markets can help to consolidate land in the hands of the
more productive farmers, perhaps by improving access to land for younger and more educated farmers, and
can provide inefficient or unviable farmers the security to seek off-farm work without fear of losing their
primary asset.




 xxviii
                                                                                              ﻿OVERVIEW




Given that most farms in India are under              The all-India weighted averages show significant
one hectare, these findings are significant—          scope for increasing yields of both crops with
politically as well as economically—for growth        technology that is already available. But India’s
and poverty reduction. The argument of “small         biophysical heterogeneity argues against taking
and efficient” still applies from a technical         such a blanket assessment at face value. Indeed,
perspective of higher yields or productivity          the states vary considerably in their potential
(that is, the value of output per acre), as found     yields, their progress in closing the yield gaps,
in the productivity analysis discussed earlier,       and the size of the remaining yield gaps. Some
but the shift in the returns to farming indicates     states have made good progress in rice in the
that “small and efficient” does not necessarily       last 15 years. Growth in wheat yields has slowed
translate to more income (profits). The main          in most states, although Gujarat, Karnataka,
reason behind this reversal is that the higher        and Maharashtra substantially narrowed their
value of output per hectare is neutralized by         yield gaps.
very high family labor costs. These trends
re-emphasize the point that family labor is           Given that no country in the world has reduced
overused on the farm.                                 its yield gap below 20 percent, 30 percent
                                                      may be a realistic target. Some states are
                                                      approaching this 30 percent target with current
Technology, Yield Gaps, and                           technology and face limited prospects for
Growth Prospects                                      further improvement. Maharashtra and Gujarat
                                                      seem to have exhausted their potential with
Many factors contribute to productivity               the current wheat technology, whereas Punjab
growth (including infrastructure, markets, and        and Haryana have limited room to increase
education). Among these factors, technology           yields. West Bengal and Punjab are close to
plays a central role by helping to increase           their potential for paddy, but most other states
yields. Changes in yields are a joint outcome         have significant scope for yield improvement.
of contributions by research and extension.           In interpreting these findings, two important
Research generates new technology that moves          caveats need to be kept in mind. One is that the
the production frontier (the yield potential)         attainable yield, used here as the benchmark for
upward and outward. Extension assists farmers         the yield gap, assumes that no biotic or abiotic
to better exploit the available technology—           stresses are present—rarely true in practice.
through access to new technology and advice to        Second, these are physical or output maximum
improve technical proficiency in using it—and         potentials, not economically optimal potentials.
in essence to close the “yield gap” (which is the     Depending on the local policy and institutional
difference between actual and potential yields).      environment, reducing the physical yield gaps
A combination of simulation models, research          may not be economically viable (as actual or
data, and actual yields helps to identify the yield   observed farm yields may more accurately
gaps and how they have evolved over time. The         reflect).
analysis focuses on rice and wheat, for which
abundant data are available, and which are an         The movement of the production frontier,
appropriate choice, given the heavy historical        or realizable yields, shows that the potential
focus on these green revolution crops.                for wheat has continued to rise as new,



                                                                                                  xxix
Republic of India: Accelerating Agricultural Productivity Growth




better-yielding varieties are released, but rice         How do yield gaps in India compare to those in
potential has stagnated. Yet between 1995                other major grain-growing areas of the world?
and 2010, yield gaps widened for wheat and               The wheat yield gap in Punjab resembles the
narrowed for rice. Growth in actual yields is            average across the other major production
similar for the two crops; the different trends          areas, with similar scope for doubling current
in their yield gaps originate in the much faster         yields. The yield gap for irrigated rice in Punjab
rise in realizable yields for wheat—36 kilograms         is larger than the global average: Yields in
per hectare per year for rainfed wheat and 54            Punjab need to rise by 75 percent to close the
for irrigated wheat—whereas realizable yields            gap, whereas on average global yields need to
for rice have stagnated since 1995, increasing by        double. The comparators for rainfed rice are
only 6 kilograms per hectare per year.                   limited, but the data show that Madhya Pradesh

Messages and Implications

Yield gaps offer an opportunity, but their persistence reflects challenges in technology services. Yield
gaps have narrowed over time, more for wheat than rice. But substantial gaps remain for both crops in rainfed
as well as the green revolution areas, signaling potential for further gains and lending weight to the need to
resolve the “extension problem.” Research too has challenges to confront: dedicating more effort to crops
other than rice and wheat, and working with a rapidly emerging private sector to align research priorities to
tackle a changing climate and to address the multiple biotic and abiotic stresses prevalent in the LIS.
Investing for innovation and change requires institutional innovation. Technology capital is a
critical input to accelerate agricultural productivity. India is among the world’s preeminent investors in
agricultural R&D. Yet success is more than the sum of public funding. The quality of innovation and the
capacity of institutions to reconfigure and reorient themselves to a rapidly transforming agriculture are
critical. To remain relevant and stimulate transformative change, the impressive institutions that ushered
in the green revolution must take action:
      In agricultural research, public investment is increasing, but a number of reforms have also been
          proposed to reconfigure the research system to meet current needs and challenges. The issues of
          relevance, efficiency, and effectiveness of the research system are well understood, but bold action
          is needed to implement the reforms. At the same time, private sector investment in research is
          rapidly increasing, with significant potential to contribute to rapid growth in productivity and
          incomes, including in the poorer semi-arid parts of the country. To fully exploit this potential, the
          remaining constraints to private investment need to be removed.
      In agricultural education, the state agricultural universities face multiple crises in fulfilling
          their mandate to build the needed human capacity for technical innovation and undertake crucial
          adaptive research and extension activities. The crises in governance, resources, effectiveness, and
          ethics is widely acknowledged. These issues are revisited in the Bhubaneswar Declaration. The
          proposed roadmap for improving India’s higher education system calls for fundamental changes but
          requires unwavering political will and commitment to overcome the ingrained resistance to change.
      In agricultural extension, the Agricultural Technology Development Agency models a
          decentralized, demand-driven approach for advisory and extension services to respond to local
          demands, priorities, and constraints. The lack of skilled, dedicated personnel in the agency, along
          with weak research-extension links, limited outreach to farmers, and limited operational flexibility,
          have yielded disappointing outcomes. Priorities are to renew and improve the focus on community
          outreach, reinforce organizational autonomy, and improve staff quality. Reforming current service
          delivery and promoting a pluralistic system is an urgent priority.




 xxx
                                                                                             ﻿OVERVIEW




rainfed yields must grow by 150 percent to close      substantial dairy surplus for export, but it is
the current yield gap.                                already a leading exporter of bovine (buffalo)
                                                      meat and a highly competitive exporter of
Certainly yield is not the only metric of             mutton and pork. Expanding markets for
research achievements. Research also aims             processed meat and halal-certified products
to develop varieties capable of withstanding          provide additional opportunities for export
specific growing conditions (evolving pests and       growth. Growth in meat output has mainly
diseases, for example) and supporting other           come at the extensive margin—from growth in
local priorities (tastes or the timing of growing     livestock numbers. Substantial scope remains
cycles, for example). Future research will need       for improvement at the intensive margin.
to emphasize varieties capable of adapting to
the pressures of climate change. For example,         More than two-thirds of livestock output is
early maturing varieties may be better adapted        from the dairy industry. India leads the world in
to rising temperatures, which are expected to         milk production, but milk yields are about half
reduce yields significantly. Varieties specifically   the world average. Formal milk processing has
suited to the needs of crop-livestock systems         expanded rapidly, driven by deregulation of the
or agro-processors will also be in increasing         dairy industry in the early 1990s. Even so, less
demand.                                               than 20 percent of milk is formally processed.

                                                      The challenges. Despite this potential,
Livestock Subsector: Significant                      India’s public expenditure on livestock is
Opportunities and Policy Priorities                   low, declining, and ineffectively targeted.
The livestock subsector has grown at twice the        The impact of public programs must be
rate of the food grain subsector. Continued           monitored to inform policy, develop a strategy
income growth and demographic change                  for the sector, and enhance service delivery.
in India will heighten demand for livestock           Most public funding goes to administrative
products and offer significant opportunities          rather than productivity-enhancing activities.
to increase production and incomes. Mixed             Allocations for the smaller species (small
crop-livestock farming systems predominate            ruminants, poultry, and pigs) that yield more
among smallholders and are an important tool          benefits for smallholders remain low and have
to target rural underemployment (particularly         declined. The limited public expenditure on
of unskilled and family labor), diversify risk,       livestock must be rationalized between public
and stabilize income throughout the year.             and private goods.
Livestock production is more inclusive than
crop agriculture, with livestock ownership more       The policy interventions. Policy interventions
widespread than land ownership. Women and             to raise productivity in the livestock subsector
other socioeconomically marginalized groups           must target services and institutions for
stand to benefit the most from better livestock       technology, marketing, and animal health. The
productivity.                                         National Dairy Plan is a multi-pronged approach
                                                      along those lines that would work for the meat
The potential. India has the world’s largest          industry as well as poultry and pig production.
livestock herd. It is unlikely to have a              Aside from encouraging commercial and



                                                                                                 xxxi
Republic of India: Accelerating Agricultural Productivity Growth




Messages and Implications

Aside from additional and better-targeted public spending, priority actions include strengthening the
institutions for technology, animal health, and smallholder marketing. The prospective socioeconomic
and environmental gains from the livestock subsector offer substantial scope for economic and green
growth. Measures that increase livestock productivity can enhance incomes, mitigate the environmental
impact of livestock production, and enable livestock to adapt to India’s challenging and changing climate.
In this context, it is imperative to rely on intensification for growth. Improvements in breeds, feeds,
and animal management and productivity are the other critical elements of any strategy to mitigate the
environmental impacts of growth in the subsector.


large-scale operations, attention must be             many areas are 5–10 times the recommended
given to technologies, institutions, and              levels, and water use per liter of milk exceeds
policies enabling smallholders to produce             the world average in most intensive and semi-
more efficiently and sustainably and compete          intensive systems. Producers must aim to
profitably in a price- and quality-conscious          maintain vegetative groundcover, reduce soil
market.                                               erosion and down-slope sedimentation, improve
                                                      water infiltration and groundwater recharge,
The states urgently require better strategies to      and increase pasture production. Greenhouse
meet local requirements for animal breeding           gas emissions per head of cattle depend
(cattle, small ruminants, pigs), improved             greatly on animal breeds and the type of feed
nutrition and feed (to realize the gains from         provided. Climate change will affect livestock
breeding), improved animal health systems             productivity through a higher incidence of heat
with better livestock support services (including     stress, drought, and flooding. A large number
disease surveillance), and policies to attract        of adaptation strategies exist, but greater and
private investment.                                   better implementation is needed.

Investments in market infrastructure and
                                                      Investments in Agriculture for
quality standards are integral to the efficient
distribution of livestock products, value             Growth and Sustainability
addition, and food safety. Much of the                Private and public expenditures to expand
investment will come from the private sector.         productive capacity. Private investment in
Aside from encouraging such investment, the           agriculture has increased rapidly since the mid-
public sector must implement measures that            1990s. It accounts for more than 80 percent
help smallholders improve their bargaining            of the investment in agriculture and largely
power and build their capacity to absorb              consists of on-farm investments, primarily
production and market risks.                          in irrigation pumps and to a lesser degree in
                                                      machinery. Overall investment on the farm
Strategy and policy for the livestock sector          remains low, and farmers prefer to allocate
must incorporate a range of environmental             more of their disposable incomes to financial
issues. Improvements in animal management             savings and expanding business capital.
and productivity are critical for mitigating
greenhouse gas emissions and relieving                Shares of public and private agricultural
pressure on land and water. Stocking rates in         investments in capital formation are declining,


 xxxii
                                                                                             ﻿OVERVIEW




however. For a sector deemed to be a priority,       Private irrigation investments, public
agriculture’s diminishing share of public            subsidies, and declining productivity. Two
investment appears contradictory. When               related trends in public and private investment
public expenditure is defined more broadly—to        are the rapid growth in private irrigation and
include expenditures in the sector (through          the large and growing volume of subsidies
agricultural programs and institutions) and          in public expenditures. The importance of
expenditures for the sector (for complementary       irrigation for agricultural growth is clear.
public investments like rural roads and              But the dominant mode of private irrigation
electricity)—it is clear that agriculture does       development—groundwater extraction—may
indeed command a much higher priority.               not be sustainable in its current form. Distorted
Based on those criteria, public expenditures         incentives encourage excessive extraction and
were equivalent to one-third of agricultural GDP     highly inefficient use of water. Power and credit
in 2009, up from about one-fifth in 1995,            subsidies are the major drivers of groundwater
about half of which was from the central             extraction. Marginal returns to these subsidies
(union) government. Infrastructure accounted         in terms of improved productivity and poverty
for the largest share (34 percent) of these          reduction were high in the 1960s and 1970s
expenditures, with input support services            but are now significantly lower than returns
following at 26 percent. Input support services      to investments in rural roads and technology
(essentially private goods and subsidies)            services.
have grown the fastest (22 percent per year).
In sharp contrast, research, education and           The political economy of subsidies in India
extension services, which have been shown            has relegated expenditure efficiency as well as
to have the largest returns among public             budgetary implications to the extreme margins
expenditures in India, had a combined share of       of public decision making. Yet concrete evidence
less than 2 percent.                                 is emerging that current policies will have an
                                                     outsized negative impact over the long run. For
Even those estimates do not capture the              example, electricity subsidies lead to an increase
totality of subsidies supporting agriculture. A      in the area planted to water-intensive crops
fuller (but still not exhaustive) accounting of      such as rice and sugarcane but also lead to a
subsidies (including food, fertilizer, irrigation,   large negative impact on groundwater levels. A
power, and others) shows that they dwarf public      fall in groundwater level by 1 meter is found to
investments, irrespective of how the latter are      reduce production of food grains by 8 percent,
defined. A vivid example is the pervasiveness        water-intensive crops by 9 percent, and cash
of subsidies in two of the largest government        crops by 5 percent. A 10 percent reduction in
programs or “missions” directed to increasing        the average electricity subsidy would reduce
agricultural productivity (the National Food         groundwater extraction by 6.7 percent.
Security Mission and the Rashtriya Krishi Vikas
Yojana). A review of their implementation in         Contrary to their intended objectives, electricity
three states shows that 88-98 percent of funds       subsidies set a vicious circle in motion,
under the NFSM and 42-87 percent under RKVY,         jeopardizing agricultural productivity over the
depending on the state, were directed toward         long run. Together with other policies such as
the provision of inputs or other private goods.      MSPs, these distorted incentives have altered



                                                                                                 xxxiii
Republic of India: Accelerating Agricultural Productivity Growth




Messages and Implications

Although the budgetary implications of the large share of subsidies have little apparent weight in
public policy decisions, policies intended to increase production and productivity clearly impose
substantial costs in terms of resource degradation and lost future productivity. Consistent with the
priority placed on agriculture, public spending has grown rapidly since the early 2000s and now equals one-
third of agricultural value added. But only a small fraction of the public expenditure goes toward expanding
the productive capacity of agriculture (to public investments or contribution to capital formation). Subsidies
dwarf public investments. The substantial costs of these choices are also being borne now. If public policies
that encourage these outcomes are not rationalized, they will further reduce productivity and returns to
farmers—outcomes that are diametrically opposite to their intent.


crop composition to favor water-intensive               1990s, but reforms have been slow, uneven,
crops, particularly in the Northwest and Mid-           and frequently reversed. When reforms are
West, which are experiencing the most severe            introduced, even partially, the private sector
groundwater crises. This relationship is evident        responds swiftly and dynamically—witness
in the strong correlation between the “virtual          the emergence of contract farming, electronic
water export” from the Northwest and public             exchanges, ICT-based market information
procurement of rice. These policies urgently demand     systems and kiosks, and myriad value chain
attention—especially in the interests of curbing the    improvements. Yet the consensus is that
mounting risks occasioned by climate change.            the marketing, trade, value addition, agro-
                                                        processing, and food safety capacity required by
Consequences of imbalanced nutrient                     a diversified, vibrant, and modern agricultural
use. The current and long-term costs of                 sector has not materialized as expected.
imbalanced fertilizer use on productivity are           Government intervention continues. Parastatals
widely recognized, but the associated changes           dominate food grain markets, and private
in land productivity may not be equally well            agricultural trade is heavily regulated.
appreciated. The evidence suggests that
productivity drops beyond a certain ratio of            The traditional chain, passing through
nitrogen to phosphorous. While the all-India            agricultural wholesale markets and traditional
median level is still below that threshold, half        urban retail, dominates the marketing of
of the farmers in the breadbasket states (Punjab        agricultural commodities, but it is inefficient,
and Haryana) are above the threshold. In other          lacks integration, and is plagued by trader
words, they have reduced their productivity             collusion in the regulated and restricted
from the peak response level. Even in Bihar             markets. Even so, alternatives—modern retail
most farmers operate on the decreasing part of          and the processing and food service sectors—
the response curve.                                     are emerging, and even traditional value
                                                        chains for staples are evolving, with changes
Marketing and Market Reform:                            in factor markets, innovations that shorten
                                                        supply chains, wider access to information
Unfinished Business                                     through mobile phones, and increasing
India has long intervened in its agricultural           downstream demand for quality and brand
markets. Regulations have eased since the               differentiation.



 xxxiv
                                                                                            ﻿OVERVIEW




Innovations to improve marketing efficiency         in effect for the most part: the Agricultural
and link small and marginal producers to more       Produce Markets (Development and Regulation)
remunerative value chains have been attempted       Act—APM(D&R) Act—and the Essential
on a limited scale, with mixed results. Firms       Commodities Act (ECA). The issues related
or other private and public entities generally      to these regulations are well recognized,
have preferred large and medium farmers for         including zoning and storage restrictions,
contract farming; established or corporate          market fragmentation and inefficiency, and
retail chains have preferred the more advanced      the requirement to sell all produce at a limited
agricultural states to the states where most        number of licensed, regulated markets in often
poor smallholders reside. The main factors          nontransparent transactions. The regulations
influencing those preferences appear to             discourage private investment in storage,
be difficulties in enforcing contracts, high        handling, and marketing infrastructure; they
transaction costs, and challenges in meeting        also constrain contract farming or direct
quality standards. Initiatives to integrate small   purchases by agro-processers and prevent
farmers to value chains failed to introduce         improvements in value chain efficiencies.
or adapt appropriate technology over time,
although producer companies appeared more           To address the major problems, the government
effective than cooperatives at linking small-       introduced a “Model Act” in 2003 and urged
scale producers to markets.                         state governments to amend their legislation
                                                    and regulations accordingly. Most states
Alternative market channels—traditional             amended their legislation, but the extent and
private sector traders, state-sponsored             implementation of reforms remain limited.
cooperatives, and Rural Business Hubs (a            An in-depth study, including field assessments
modern private sector innovation)—also show         in five states, finds a largely unfinished
very mixed results. By and large, the benefits      agenda. Despite attempts to modify the ECA,
and costs differed little among the alternatives.   its essential provisions remain intact. The
The alternatives showed no clear improvement        multiplicity of control orders issued by multiple
in smallholders’ access to inputs or outputs, and   agencies (at the central and state levels) creates
smallholders experienced no significant price or    uncertainty and raises transaction costs. The
quality discrimination in any market channel.       lack of transparency prevails, restricting trade
What emerges is that medium and larger              and maintaining market segmentation.
farmers have better access to state-sponsored
cooperatives, indicating that subsidies (either     Different states interpret and implement
through access or merely through the scale          reforms in different ways, prompting a need
effect) are not as pro-poor as intended.            to rethink market governance. Agriculture
Cooperatives are more beneficial in the more        is a state subject, but inter-state trade and
remote or backward areas of the states studied,     commerce is under the Union list. For
where the private sector is thin or nonexistent.    efficient markets, it is critical to establish
                                                    common norms and ensure transparency
This evidence from the ground level provides        and predictability of rules and regulations to
a useful backdrop to two pervasive regulations      facilitate private trade and investment. While
that have “stifled” agriculture and remain          overregulation is clearly a hindrance, the other



                                                                                                xxxv
Republic of India: Accelerating Agricultural Productivity Growth




Messages and Implications

In most cases, market reforms have not been implemented in full or in earnest, with provisions
and omissions that effectively retain the status quo and restrict private investment and trade. India
remains a segmented agricultural market; reforms to lift restrictions on movement, stock, and trade remain
limited. Different states interpret and implement reforms and regulations in different ways, suggesting a
need to rethink market governance. It may be advisable to place agricultural marketing on the Concurrent
List and establish common norms for taxation and other charges/fees to make the system more transparent
and predictable.
State governments face a conflict of interest with the proposed market reforms. Reform of APMCs have
fiscal implications for state government, and as such have revenue implications. To overcome this obstacle,
a cost-sharing strategy will need to be devised. Another conflict of interest is that state departments or
marketing boards both run and regulate markets. An independent market regulator could level the playing
field between the state and private markets.


extreme of an absence of any regulation is also        industry has a dualistic structure, with a
not desirable. Markets need to be orderly and          relatively small (in number of units) but capital-
governed well to create a healthy environment          intensive organized segment coexisting with
for transactions, calling for an independent           a pervasive, mostly rural, and more labor-
market regulator.                                      intensive unorganized segment. Rural firms
                                                       are less capital intensive, less productive, and
                                                       dominated by small family enterprises.
Beyond the Farm: Exploiting the
Potential for Food Processing                          Shares of food processing in all manufacturing
                                                       in terms of employment and numbers of units
Urbanization, the shift to nonfarm activities,         have been stable in most states over time,
rising incomes, women’s increasing participation       while the share of output has increased. Both
in the labor force, and changing consumption           the organized and unorganized segments have
habits will create an outpouring of demand             experienced capital deepening and declining
for processed foods. Food processing provides          labor intensities. The organized segment has
a natural entry point for India’s sluggish             generated jobs with an increased number of
“manufacturing” sector to move into                    units, but the unorganized segment has lost
predominantly agricultural areas and create            enterprises and jobs over time. The remaining
much-needed off-farm employment. Food                  enterprises are larger in scale, with rapidly
processing has among the largest multiplier            growing output per unit and slower but still
effects across the economy. It can stimulate           growing employment per enterprise.
higher agricultural productivity through better
and more stable farm prices, reduce wastage by         Labor productivity has risen fast, keeping
transforming produce unsuitable for wet markets        pace with labor productivity in the non-food
into value-added consumables (increasing returns       sector, but the associated rise in wages (also
to farmers), and promote diversification.              relative to the non-food sector) has restrained
                                                       growth in employment, encouraging further
Food industry structure and investment.                capital intensity and scale of operating units.
Like all manufacturing, the food-processing            The preference for higher capital intensity


 xxxvi
                                                                                                     ﻿OVERVIEW




and labor-saving technology reflects the                 by starting small nonagricultural enterprises
perpetuation of informality and a reluctance to          or operating such enterprises alongside
hire labor, perhaps due to labor laws and other          their agricultural ventures for supplemental
factors in the business environment.                     income. Third, industrial investments in
                                                         individual states are highly concentrated in
The significance of these trends for employment          specific sectors. Among the factors explaining
and transformation, even in the more rural               investment patterns, the main results of
food-based industry and in the more populous             interest for policy are backward linkages to
and poorer states, cannot be overstated.                 agriculture, credit, and public infrastructure.
Encouraging new businesses to enter and                  The food industry’s concentration in
existing businesses to expand employment                 agricultural states clearly suggests that “location
more rapidly than in the past will create                matters.” Locations with higher agricultural
employment, but only with greater attention              productivity attract more private investment
to the enabling environment and barriers to              and employment in food processing. Finally, the
entry for smaller firms, especially in the lagging       findings on access to credit and infrastructure
states.                                                  reconfirm their well-known importance as
                                                         determinants of investment.
Patterns and drivers of private investment
in food manufacturing. Three major findings              Productivity growth in the food-processing
emerge from the analysis. First, the organized           industry. TFP growth (TFPG) rebounded sharply
food industry is more prevalent in less                  after 2000, following a decline in the 1990s. For
industrially developed states with higher shares         the unorganized segment, TFPG was positive
of income from agriculture; the unorganized              but much lower than in the organized segment.
segment is more prevalent across states,                 The strong, positive correlation between TFPG
regardless of the level of development. Second,          for the organized and unorganized segments
food manufacturing is more dominant in                   across states indicates that states doing
states with a higher percentage of poor people.          better in one segment also do better in the
Individuals tend to diversify out of agriculture         other, suggesting a better overall investment

Messages and Implications

A dual focus on agricultural productivity and creating an appropriate investment climate to attract
private investment is needed to promote more rapid rural transformation. States where agriculture is
dominant have significant potential to attract private investment in the food industry. Higher agricultural
productivity also attracts private investment in food processing, placing a premium on supportive
investments in (for instance) irrigation, roads, and better functioning markets. Active promotion of this
relatively more labor-intensive industry, through appropriate incentives, in states with low per capita
income, high dependence on agriculture, and high incidence of poverty will help create off-farm jobs, which
will absorb more people from agriculture and promote structural transformation. These arguments are
even more relevant with respect to the unorganized segment of the industry, which is more widespread and
has higher potential to absorb labor but currently suffers from low productivity. High levels of efficiency in
many states signal that the food industry is well placed to compete in a more liberalized marketplace, with
potential scope for foreign direct investment as well as exports. Policies to improve the investment climate
for agro-processing, both for the organized and unorganized segments, is thus a high priority.



                                                                                                        xxxvii
Republic of India: Accelerating Agricultural Productivity Growth




climate for food processing in some states. Key     and livestock yields are typical, with large
determinants of productivity differences across     differentials across districts. Rapidly diversifying
states include backward linkages to agriculture,    production has translated to faster growth
infrastructure, and investor friendliness. Beyond   and poverty reduction. Despite these seeming
the backward linkages, other agglomeration          similarities, the two states have experienced
economies appear not to be significant.             distinct growth patterns. Diversification was
                                                    strong and consistent in Bihar, with prices
The unorganized segment does not                    playing a lesser role in growth. Improved yield
appear to grow through a complementary              spurred growth in Odisha in the early 2000s,
relationship (for example, through                  but now growth is led primarily by prices. As
outsourcing or subcontracting) with                 noted earlier, price-led growth raises concerns
organized food manufacturing. Both the              about its sustainability, given that yields are
organized and unorganized segments seem             declining. In contrast, growth in Bihar appears
to be operating below optimal scales. The           more robust.
significantly suboptimal scale of operation
in the unorganized segment suggests strong          Small, fragmented holdings prevail in both
disincentives for employing labor in an             Bihar (72 percent are under 0.5 hectare) and
otherwise efficient industry.                       Odisha (60 percent smaller than 1 hectare).
                                                    Tenancy and sharecropping arrangements
                                                    discourage investment, while fragmented
Transforming Agriculture in LIS:                    holdings and the small scale of operations
Challenges and Priorities for Bihar                 are a drag on efficiency and constrain output
and Odisha                                          marketing. Inevitably, transaction costs and
                                                    market risks tend to be high.
The eastern states offer enormous agricultural
opportunities if their natural resources are        Low input use intensity and efficiency limit
managed judiciously, in a framework of              productivity growth, although fertilizer use has
appropriate policies and institutions and           grown faster than the national average in Bihar,
supportive infrastructure. Sufficient water,        approaching the levels in more advanced states.
a suitable climate, and significant scope to        As in those states, fertilizer use is imbalanced,
improve yields make the eastern states a            taxing productivity. The high level of biotic
valuable resource for sustaining national food      and abiotic stress from many sources—pests,
security, and their potential for high-value        diseases, drought, floods, acidic and sodic
horticulture and livestock production portend       soils—means a very high level of agricultural
a rapid economic transformation. Analysis in        risk, and likely constrains the adoption of new
two of the poorest states, Bihar and Odisha,        technology, reduces the efficiency of input use,
reveals the challenges and priorities involved in   and discourages productive investments.
realizing this scenario and contributing to the
inclusive growth agenda for LIS.                    Marketing remains a challenge. Relatively low
                                                    and volatile prices result from thinly spread
Challenges to productivity and                      and underdeveloped markets; market density
diversification. In both states, low crop           is low, and few markets have sophisticated



xxxviii
                                                                                                   ﻿OVERVIEW




infrastructure such as cold storage facilities.         irrigation and infrastructure than other states.
Bihar’s bold abolition of the APM(D&R) Act              Allocations remain lower than the all-India
seems to have unintentionally left a void               average—inconsistent with the objective of
in market governance, with allegations of               inclusive growth. Both states improved road
noncompetitive behavior. Markets have been              density, but other infrastructure deficiencies
taken over by private entities; farmers continue        persist, particularly in rural electrification.
to pay market fees but receive no services,             Farmers invested in the more costly diesel
and public and private investment in market             pumps in Bihar, where irrigation intensity
infrastructure is nil. In Odisha the Act is intact,     (at 62 percent of net sown area) now surpasses
but most markets still have no mechanism for            the all-India average (45 percent). Odisha’s
price discovery and price determination.                irrigation intensity remains low. Bihar
                                                        has significant potential for surface water
Bihar, Odisha, and other LIS historically               irrigation; Odisha already sources irrigation
received much less public funding for                   water from canals.


Messages and Implications

The top priorities for Bihar and Odisha are to tackle the multiple biotic and abiotic stresses, develop
markets, and promote livestock development. Small holdings make it a priority to enable people to
leave agriculture or diversify into higher-value agriculture. Technological solutions range from sustainable
crop management strategies (for overcoming soil acidity and sodicity, multi-cropping, reducing water use,
and conserving other resources) to better seed quality and faster seed replacement, and research on abiotic
stresses (together with private companies).
Collective action—through producer organizations, cooperatives, farmer associations, or self-help groups—
is important for scale economies and linking smallholders to value chains. Bihar has successful approaches
that can be scaled up (for dairy, vegetables), and Odisha is innovating with farmers’ markets and cluster
approaches for specific horticultural crops. These initiatives need to be set within an overall framework
conducive to private investment in marketing, agro-processing, and land development. Establishing a
regulatory framework for fair, transparent, and efficient markets with free movement of goods and services
is essential.
Livestock is a “quick win” for inclusive growth, nutritional outcomes, and employment. Priority actions
include focusing research on diseases and production constraints; policy and other support for small
ruminant, pig, and poultry production; and incentives for private investment in processing, value chains,
feed production, and veterinary services.




                                                                                                       xxxix
                       A. Context and Objectives of this Report

1                      Introduction to the Issues
                       and Analytical Approach




N
        o longer a “sleeping giant,”1 India has at             today than in the 1960s, despite the population
        last shaken off the persistent “Hindu”                 growing by two and half times.
        rate of growth2 (averaging about 3.5
percent per year from 1950 until the mid-1980s)                With the economic reforms of 1991, gradual
to attain sustained GDP growth of 8 percent per                liberalization in the agricultural sector started
year in the last decade (2001–11). The timing                  in the Eighth Five Year Plan (1992–97), with
and factors contributing to this structural break              successive plans increasingly emphasizing
from historical growth rates are vigorously                    reforms in accord with evolving policy
debated,3 but the turnaround in growth since at                directions. Table A.1.1 (Annex 1) highlights
least the early 1990s is undeniable.                           some of the major policy reforms and other
                                                               developments affecting the sector since 1990/91
Indian agriculture has also undergone some                     and some of the private sector response to these
profound transformations. Food security                        changes. These incremental policy shifts, along
has been an overriding priority in India’s                     with the trade-related reforms starting in the
development strategy since the 1960s,                          mid-1990s, set in motion other transformations.
underpinned by a comprehensive and directive                   Once subject to heavy implicit taxation and
policy, public investment and institutional                    other policy discrimination, agriculture now
support through a sharply focused program for                  benefits from a heavily supportive (heavily
urgent scale-up of the (then) new but relatively               subsidized) policy regime. India has emerged
simple cereal production technologies. This                    as a global agricultural powerhouse—it is the
strategy paid off with India’s impressive                      second-largest producer in terms of aggregate
transformation from a chronic food-deficit                     agricultural produce, as well as a top producer
and food-aid-dependent country into a food-                    of several major commodities.4
surplus country that is now a net food exporter.
Growth in the production of staple foods (rice                 Since the early 1990s, India’s agricultural
and wheat) has outpaced population growth.                     exports have also steadily grown and
Per capita output of cereals is 20 percent higher              diversified, to the extent that its nontraditional
                                                               exports now dominate the traditional exports
1	 Bhagwati (2002).
2	Virmani (2006).
3	 See, for example, Virmani (2006), Basu and Maertens         4	   India ranks among the top two or three producers globally of
   (2007), Bosworth, Collins and Virmani (2007), and Kotwal,        wheat, rice, pulses, oilseeds, cotton, sugarcane, tea, milk, and
   Ramaswamy, and Wadwha (2011).                                    fruits and vegetables.




                                                                                                                              1
Republic of India: Accelerating Agricultural Productivity Growth




of tea, coffee, spices, and cashews.5 Growth                          with the urbanized, highly skilled and outward-
in some of the nontraditional exports reflects                        looking service sector, propelled India into
sustained efforts by the government at export                         the ranks of the world’s fastest growing
promotion and in some cases the introduction                          economies. But it exists alongside the largely
of new technology.6 The performance of exports                        agricultural and unskilled rural Bharat, which
is all the more impressive considering that                           seems fixed on a path of low long-term growth
agricultural trade policy, while relatively more                      fluctuating around a rate of 3 percent per year.
liberalized than before, has remained restrictive                     Agricultural growth has picked up in the past
and unpredictable, with frequent changes                              decade (2002/03 to 2012/13) to about 3.5 percent
(Gulati, Jain, and Hoda, 2013).                                       (trend rate), almost as high as the previous
                                                                      peak reached in the mid-1990s, but it remains
Rapid income growth following the economic                            volatile and below potential. At present, the
reforms in the 1990s and some key policy                              elements of dynamism noted previously do
changes attracted significant private interest                        not appear sufficiently strong to influence
and investment, encouraging a number of                               agriculture’s overall performance.
potentially transformative developments.7 The
most visible and dynamic changes have been the                        Progress on inclusiveness has been less than
spread of Bt cotton in western India, soybeans                        expected,8 and inequality is on the rise (World
in the center, poultry in the southern areas, and                     Bank 2010). The rural-urban income divide
maize in the eastern areas, as well as remarkable                     is widening, with the key indicator of labor
innovations in e-trading, internet kiosks, and                        productivity showing growing differentials. In
other value chain and marketing tools.                                parts of the country, persistent poverty and
                                                                      farmer suicides appear to signal a pervasive
Despite the headline figures on growth, the                           agrarian crisis (Dev 2008a, 2013a; Mishra
increased elements of dynamism in agriculture,                        2008; GoI 2006a). These facets of the economy
and India’s export performance, concerns linger                       are major concerns for the government, as
about the pattern of growth. India continues                          reflected in the new Twelfth Five Year Plan’s
to face daunting challenges in reducing the                           development objective: “Faster, More Inclusive,
number of poor, curbing high malnutrition, and                        and Sustainable Growth.”
achieving shared prosperity. Economic dualism
persists. Modern India, popularly associated                          The structural transformation of the economy
                                                                      has not progressed as fast or as widely as may
5	   In value terms, India’s agricultural exports have consistently
     exceeded agricultural imports. India has emerged as the          be expected, and economic growth has been
     largest exporter of rice and beef (buffalo) meat in 2011–12.     characterized as “lopsided” (Dev 2008b). With
     See Gulati, Jain, and Hoda (2013).
6	   The Agricultural and Processed Food Products Export
                                                                      the majority of India’s labor force still in
     Development Authority (APEDA) was established in 1985 with       rural areas, and over 68 percent employed in
     a mandate to develop and promote exports of agricultural
     and processed food products from India. It has successfully
                                                                      agriculture, India’s development aspirations
     engaged in sustained efforts to improve the transparency of      must inevitably reckon with agriculture’s
     supply and institute the traceability and quality standards
                                                                      political, social, and economic importance.
     required to negotiate entry into the European Union market
     for Indian grapes as well as the Japanese and United States
     markets for fresh mangoes.                                       8	   Prime Minister’s speech to the Full Planning Commission
7	   Three key policy changes were to allow FDI in wholesale               meeting to discuss the “Issues for Approach to the 12th Five
     trade (1997), the de-licensing of Food Processing Industries          Year Plan” (see the Planning Commission website,
     (1999), and the New Seed Policy (1998).                               http://planningcommission.nic.in/).




     2
                                                        Chapter 1: Introduction to the Issues and Analytical Approach




Indeed the Eleventh and Twelfth Plans explicitly                      nonagricultural and non-casual sectors, such as
set a target of 4 percent for agricultural growth                     manufacturing (World Bank 2012a; GOI 2013a).
to achieve their overall economic growth
targets and to ensure that growth is sufficiently                     Going forward, the challenges of sustaining food
inclusive.9 Geographical inclusion—to focus on                        security and promoting shared prosperity seem
LIS and rainfed areas—and social inclusion—to                         more difficult now than in the past. Growth
respond to the feminization of agriculture and                        and food security will need to be sustained
level the playing field for the disadvantaged                         with fewer degrees of freedom. The quality of
social groups who are a substantial part of the                       land and water is declining, and agriculture
agricultural economy—are well-recognized                              faces intense competition for land and water
development priorities.                                               from urbanization and nonagricultural uses.
                                                                      The sharp, sustained focus on food production
Agriculture’s weak performance affects the                            helped India achieve self-sufficiency, but the
livelihoods of a disproportionately large                             complex web of policies and institutions
number of people, contributes to the slower-                          that emerged in the process is now argued to
than-desirable decline in poverty,10 and sustains                     constrain more robust, sustainable growth
the perception that national food security is                         (Gulati, Ganguly, and Shreedhar 2011), limit the
under threat despite record harvests of the                           performance of markets (Gulati, Landes, and
main staples.11 Equally vexing is the recent                          Ganguly 2009), and discourage much-needed
phenomenon of seemingly “growth-less jobs”                            diversification to accommodate changing
(Dev 2013b)—a rise in rural employment                                consumption patterns (Joshi and Gulati 2007;
(as inferred from rising real farm wages)                             Joshi, Gulati, and Cummings 2007).12
alongside sluggish agricultural productivity
growth—which compounds the more widely                                Weather—especially the vagaries of the
recognized challenge of “job-less growth”                             monsoons—remains a major source of volatility
(Bosworth and Collins 2008), referring to                             in Indian agriculture. Heavy public investments
lack of “good jobs” in the more productive,                           and the green revolution technology helped
                                                                      India achieve its food security goals, and they
9	  For the Eleventh Five Year Plan, the Planning Commission          have made India the world’s most intensively
    estimated that 4 percent growth in agriculture was necessary      cultivated and most irrigated country (in
    to achieve the overall GDP growth target of 9 percent without
    undue inflation and to meet its objective of inclusiveness (GOI   terms of area equipped for irrigation). Those
    2011a). Similarly, for the Twelfth Five Year Plan, the Planning   investments have helped stabilize agricultural
    Commission has again set a goal to achieve agricultural growth
    of at least 4 percent to meet its targeted GDP growth of 9–9.5    growth, shielding it to some extent from the
    percent and ensure that this growth is inclusive (GOI 2011b).     worst effects of unpredictable weather,13 yet
10	 Datt and Ravallion (2009) find that rural growth has
    contributed relatively less to poverty reduction in the 1990s,
    compared to their earlier findings for the 1980s, when it was     12	 Since the initiation of the green revolution (with its focus on
    a more significant driver of poverty reduction than                   technology and water), agriculture has been characterized
    urban growth.                                                         by a high degree of public support aimed at increasing the
11	 Revised estimates by the Expert Group on Methodology                  use of productivity-enhancing inputs (seed, fertilizer, water,
    for Estimation of Poverty, chaired by S. D. Tendulkar, put            and electricity), external trade controls, zoning and storage
    the rural poverty headcount at 42 percent in 2004/05, up              restrictions, output marketing regulations, and a vast system of
    from the previous estimate of 28 percent and closer to the            food price policy and public procurement and distribution of
    estimate of the proportion of people living under US$ 1.25            food grains put in place to promote food security for the poor.
    in international purchasing power parity terms. See the           13	 Government of India Planning Commission, Twelfth Five
    Planning Commission website, http://planningcommission.               Year Plan 2012–17, http://planningcommission.gov.in/plans/
    nic.in/eg_poverty.htm.                                                planrel/12thplan/welcome.html.




                                                                                                                                    3
Republic of India: Accelerating Agricultural Productivity Growth




the target of 4 percent growth has remained                           strategy adequately prioritizes actions to create
elusive. Slowing growth rates in the late 1990s                       the conditions for agriculture to achieve and
and early 2000s14 and food price increases in                         sustain faster growth.
more recent years are stark reminders of threats
to food and nutritional security. As the effects of                   The scope and approach for this study were
climate change become more pronounced, they                           arrived at through a consultative process,
will likely magnify the threats to food security                      which deliberately guided the study to focus
and growth and compound the challenges of                             on issues related to productivity growth, the
developing new agricultural technology and                            central challenge facing Indian agriculture.
managing scarce land and water resources.                             India is a large and heterogeneous country, with
                                                                      widely varied agro-ecological and economic
                                                                      endowments. A comprehensive study that can
Motivation and Scope of this Study                                    do justice with sufficient breadth and depth
Agriculture’s persistent low growth (especially                       to all of the complex issues surrounding its
its low productivity) jeopardizes achieving                           agricultural development would be an enormous
the goals of more inclusive and sustainable                           undertaking. This study is less ambitious, focused
economic growth. The imperative of improving                          on the specific issue of productivity growth and
agricultural productivity is widely recognized.15                     cognizant of the wealth of existing knowledge,
It is reflected in efforts to reverse the perceived                   completed studies, and ongoing analytical efforts
neglect of agriculture in the Eleventh Five                           (some of which address other major pressing
Year Plan with the allocation of substantial                          issues, such as water and irrigation). Finally,
resources to kick-start the sector.16 The                             the study takes a deliberate empirical and
government adopted a fairly comprehensive                             evidence-based approach, to objectively inform
strategy targeting technology, irrigation and                         policy and institutional debates. Inevitably, lack
water management, agricultural diversification,                       of or insufficient quality of data, and at times
infrastructure, and private investment in                             challenges in obtaining data, prevent a deeper
marketing and food processing. The central                            dive into some pertinent issues.
question examined here is whether the current
                                                                      For these reasons, it is important to
14	 The deceleration was widespread, affecting all categories         acknowledge that this study is not intended
    of farmers (small and large) and virtually all regions
    of India, and it affected not only food grains but other          as an exhaustive exploration of all major
    subsectors, such as livestock and horticultural crops             issues affecting agriculture. Some of the issues
    (Bhalla and Singh 2001, 2010).
15	 Despite the Indian economy’s rapid growth and changing            have been well analyzed or are being studied;
    structure, recent research shows that agricultural yields         others are emerging and more complex, and
    continue to have a substantial impact on poverty (see
    Himanshu et al. 2010).                                            their full analysis requires more focused and
16	 The core challenges identified by the Steering Committee          dedicated efforts. The former include, for
    on Agriculture for the Eleventh Five Year Plan are long term
    in nature. They include increasing population pressure;           example, irrigation and water management,
    technology fatigue (no major breakthroughs in agricultural        food grain management, and the analysis of
    technology); inefficient use of the available technology,
    owing to a weak extension system; falling groundwater
                                                                      public expenditures. The latter include food
    levels and inefficient use of water; degradation of the natural   safety and standards, agricultural trade, and
    resource base; no expansion in irrigated area; inefficient and
    distorted input markets (seed and fertilizer); and inefficient
                                                                      food price increase, all of which are high
    output marketing systems.                                         priorities for the future analytical agenda.



    4
                                                       Chapter 1: Introduction to the Issues and Analytical Approach




Nevertheless, the longer-term strategic issues                     To guide the identification of binding constraints,
related to productivity undoubtedly interact                       the diagnostics dig deeper into the drivers of
with a number of contemporary issues, such                         productivity growth, analyzing the relative
as food prices, and vice-versa. Several of these                   contributions of technological progress and
issues are naturally touched upon and discussed                    efficiency, with further insights into the technical
in context, but without the deeper diagnostics                     and economic aspects of efficiency improvement.
to fully inform the surrounding policy and                         This helps provide insights into whether
institutional actions needed to address them.                      technological progress, evident in the growth
                                                                   in physical food output over the years, is also
                                                                   contributing to income growth, and ultimately
Specific Objectives and Analytical                                 to greater prosperity in farming communities.
Approach                                                           In reality, a combination of factors is at play,
To identify priorities for accelerating                            and a rigorous approach helps to systematically
sustainable agricultural growth, this study                        identify the different sources of growth and the
specifically sets out to examine patterns of                       factors that may be driving or constraining their
productivity growth, identify the drivers of past                  contributions to overall productivity.
and current productivity growth, and outline
the binding constraints to future growth.                          The second important aspect of the diagnostic
                                                                   framework is the unit of analysis. Most analyses
To better analyze which policy, institutional,                     have relied on national or, to a limited extent,
and investment factors constrain more rapid                        state-level, secondary data, which are reasonably
agricultural growth, this study attempts to take                   abundant in India, although not necessarily easy
account of the agro-ecological heterogeneity of                    for researchers to access. National averages,
India to identify the main drivers of productivity                 and to a relatively greater extent state averages,
growth. For example, productivity growth can                       are useful to assess overall progress, but they
result from higher yields (for crops or livestock)                 are not particularly informative in developing
or increased diversification into higher-                          a strategic approach when considerable
value crop or livestock outputs (by individual                     heterogeneity is present. Because agriculture
households or across geographical areas that                       has performed significantly differently in
still focus on traditional or subsistence crops).17                different parts of India, the question arises as to
Subject to the data available, the diagnostics                     whether this variation results from differences
rely on different measures of productivity—total                   in access to appropriate technology, exogenous
factor productivity as well as land and labor                      agro-ecological conditions, lack of economic
productivity, broadly defined as the aggregate                     (including market) and physical infrastructure,
value of output per unit of land or labor.                         governance, or other reasons. The limited
                                                                   analyses available on these deeper aspects of
                                                                   agricultural performance results partly from a
17	 Improved incentives through higher prices are well-            surprising scarcity of or lack of access to reliable
    established determinants of farmers’ adoption of
    productivity-enhancing technologies and investments.
                                                                   subnational data.
    Clearly inflation is not desirable, but an improved share of
    farmgate prices in final consumer prices through improved
    marketing efficiency will contribute significantly to farm
                                                                   This study seeks to overcome those limitations
    productivity and incomes.                                      by using two relatively new datasets that



                                                                                                                  5
Republic of India: Accelerating Agricultural Productivity Growth




can provide valuable insights into Indian          Despite the progress achieved so far, large parts
agriculture. One dataset, from the International   of the country remain much less developed.
Crops Research Institute for the Semi-Arid         Formerly termed the “Lagging States” and now
Tropics (ICRISAT) and NCAP, is a compilation       referred to as “Low-income States” (LIS), these
of key district-level variables from a myriad of   generally resource-poor areas depend primarily
agencies, ministries, state level databases, and   on rainfed agriculture. They show considerable
other sources. The second dataset was collected    unexploited potential for productivity growth
by NCAER over three rounds of national             (often equated with cereal yield gaps in
household surveys conducted since 1982.            keeping with the concerns about food security).
                                                   Technology could unleash that potential for
Finally, to complement these quantitative          cereals and other agricultural products, but
analyses, qualitative aspects of the enabling      the much needed complementary public
environment—such as local institutional            investments have long bypassed these areas.
and governance factors, or specific physical
endowments that dictate which technological        Diversification into higher-value crops and
and economic options are available—are             livestock products—an important source of
assessed through in-depth case-studies (with       productivity and income growth in a rapidly
specific focus on agricultural extension,          developing economy—is taking place almost
marketing and the tailored strategies to           across the entire country, yet the pace has been
promote productivity in two lagging states,        insufficient to provide an appreciative boost
Bihar and Odisha).                                 to agricultural growth. As a result, the supply
                                                   of high-value commodities has not kept pace
                                                   with demand, spurred by rising incomes and
The Broader Context and Structure                  urbanization. The result has been higher prices
of This Report                                     and increased pressure on inflation.

A broad characterization of agricultural           Following this introduction (Part A), the study
performance that emerges from the literature       is organized into four analytical blocks, starting
and strategy documents is that parts of the        with an assessment of the performance of
country, specifically the irrigated rice- and      Indian agriculture (Part B). Part C contains
wheat-producing areas that benefited greatly       extensive and detailed analyses of the sources
from green revolution technology, appear           of growth in agricultural productivity (including
to be showing diminishing returns to that          the livestock subsector). Part D reviews issues
technology. This “technology fatigue” suggests     beyond the farm gate, focusing on investments,
that the scope for further growth in these areas   agricultural markets and the food processing
is limited in the absence of major technological   industry. Given that an important element of
breakthroughs. At the same time, pervasive         the government’s broader strategy is to promote
environmental degradation in these same            growth in agricultural productivity in the
areas threatens the hard-won gains they have       eastern states, where the potential to transform
achieved, suggestive of a “policy fatigue”         agriculture and reduce poverty is substantial,
as past policies may now warrant a fresh           Part E identifies priorities for achieving those
perspective.                                       objectives.



   6
                         Performance of Indian Agriculture: Temporal,
                      B. 


2
                         Spatial, and Comparative Perspectives
                      The Context and Conundrums
                      of Contemporary Agriculture




I
    ndian agriculture has been widely studied.18                          Per capita availability and production of
                                                                Figure 1: 
    Without going into a detailed review, this                            cereals (grams/day)
    section highlights salient features of the                  550
current agricultural economy to provide an                      500
understanding of the background and context
                                                                450
for the analyses that follow.
                                                                400

                                                                350
India’s Food Security Puzzle                                    300
                                                                   1960
                                                                   1962
                                                                   1964
                                                                   1966
                                                                   1968
                                                                   1970
                                                                   1972
                                                                   1974
                                                                   1976
                                                                   1978
                                                                   1980
                                                                   1982
                                                                   1984
                                                                   1986
                                                                   1988
                                                                   1990
                                                                   1992
                                                                   1994
                                                                   1996
                                                                   1998
                                                                   2000
                                                                   2002
                                                                   2004
                                                                   2006
                                                                   2008
                                                                   2010
                                                                   2012
For India, the priority for agricultural policy and
strategy historically has been to achieve food                                    Availability   Production
security. The performance of agriculture is thus
most often assessed by Indian policy makers                     Source: Authors, using DES, MOA data.
in terms of its success in achieving national
food security—essentially self-sufficiency in                   has increased gross cultivated area, but the
food, specifically cereals (rice and wheat). By                 prospects to bring new land into cultivation
this yardstick, Indian agriculture has achieved                 appear limited (Annex 2, Figure A.2.2).
significant success, given the rising per capita                Following a sluggish decade from about the
production of cereals between the 1960s and                     mid-1990s, production growth appears to be
now (Figure 1).                                                 back on track.

The rapid increase in cereal production                         Figure 1 also shows how per capita cereals
seen throughout the 1970s and 1980s                             availability (defined as production net of change
came from growth in yields driven by the                        in stocks and trade) has declined in recent years.
green revolution; net cultivated area has                       After rising in the 1980s and 1990s, cereals
remained largely unchanged since the 1970s                      availability has fallen to levels near those
(Annex 2, Figure A.2.1). Irrigation development                 of the 1960s, causing concern among policy
                                                                makers (GOI 2013a), who are also concerned
18	 For technical background and discussions of some            by the persistence of hunger at a time when
    contemporary issues in Indian agriculture, see GOI (2007,   food production is at an all-time high. This
    2012); Dev (2008); Gulati and Ganguly (2010); Kotwal,
    Ramaswamy, and Wadwha (2011); Ferroni (2013); and Chand
                                                                disparity is as much of a puzzle as the income-
    and Bajar (2012).                                           calorie puzzle—the consistent decline in the



                                                                                                               7
Republic of India: Accelerating Agricultural Productivity Growth




           ercentage change in per capita
Figure 2: P                                                                    Food prices relative to prices of all
                                                                     Figure 3: 
          consumption (kg/l/no.), 1983/84–2009/10                              commodities
120                                                                  140
                                                                     130
100
                                                                     120
 80                                                                  110
                                                                     100
 60
                                                                      90
 40                                                                   80
                                                                      70
 20




                                                                           1970-71
                                                                           1972-73
                                                                           1974-75
                                                                           1976-77
                                                                           1978-79
                                                                           1980-81

                                                                                          1982-83
                                                                                          1984-85
                                                                                          1986-87
                                                                                          1988-89
                                                                                          1990-91
                                                                                          1992-93

                                                                                                         1994-95
                                                                                                         1996-97
                                                                                                         1998-99
                                                                                                         2000-01
                                                                                                         2002-03
                                                                                                         2004-05
                                                                                                         2004-05
                                                                                                         2006-07
                                                                                                         2008-09
                                                                                                         2010-11
                                                                                                         2012-13
  0

-20                                                                        Base 1970-71   Base 1981-82   Base 1993-94 Base 2004-05

-40

          Pulses       Cereals             Sugar             Milk    Source: 
                                                                             Authors, using RBI online database. Ratio of Food
          Edible Oil   Meat, Fish & Eggs   Fruits & Vegs.                    Articles to All Commodity indices, Annual Average
                                                                             WPI, preserving the base years for each time
Source: NSS data 1983/84, 2009/10, national aggregates.                      period.


consumption of calories seen since the 1980s,                        Persistent Food Price Rise
across all income categories, despite rising
incomes.19                                                           The previous discussion suggests that food
                                                                     insecurity may stem more from a lack of access
Since the 1980s, Indian diets appear to have                         to food than from the unavailability of food.
become more diverse (Figure 2). Per capita                           One explanation for this outcome could be the
consumption of grain (cereals and pulses) has                        persistent increase in food prices compared to
fallen as consumption of other commodities                           prices of non-food commodities for most of the
has increased. This trend is consistent with                         1980s to the present, except for a brief period
theory, which suggests that the consumption                          the mid-2000s (Figure 3).
of low-income-elasticity commodities declines
as incomes rise.20 The “puzzle” is that calorie                      A significant share of the food price increase
intake has fallen consistently, even at low levels                   in recent years has been driven by high-value
of income and in both rural and urban areas,                         agricultural commodities, reflecting rising
despite the high incidence of undernutrition—                        demand with changing dietary preferences
that is, a basic calorie deficiency, without even                    as incomes and urbanization increase. This
accounting for the full extent of malnutrition.21                    so-called “protein inflation” is thought to
                                                                     reflect a structural problem (Mohanty 2011)
19	 See Deaton and Dreze (2010); Gaiha et al. (2012).                caused by a lag in the supply-response for the
20	 Using a quadratic AIDS demand model, which allows a
    nonlinear expenditure effect on demand, Ganesh-Kumar             commodities increasingly in demand (Dev
    et al. (2012) estimate expenditure elasticity for rice, wheat,   2013a). The components of the Wholesale Price
    and pulses to be negative, while they are positive for other
    commodities and greater than 1 for livestock products (eggs,     Index (WPI) confirm that high-value agricultural
    chicken, meat, and fish). Similar estimates were obtained by     commodities (fruits, vegetables, and milk and
    Kumar et al. (2011).
21	 The debate continues on the level and trends in calorie
                                                                     other livestock products) account for the bulk of
    consumption in India.                                            food price rise (Figure 4).



      8
                                                 Chapter 2: The Context and Conundrums of Contemporary Agriculture




Figure 4: Contribution to food price rise (percent)                              Figure 5: Real Minimum Support Prices

50%                                                                                                      1000

40%                                                                                                      900




                                                                                2004/05 Rupees/Quintal
30%                                                                                                      800

20%
                                                                                                         700
10%
                                                                                                         600
 0%
                                                                                                         500
-10%
        Grains




                                   Animal Pdts



                                                    Conds/Spices



                                                                   Other food
                         Milk
                 F&V




                                                                                                         400

                                                                                                         300




                                                                                                                1975
                                                                                                                1977
                                                                                                                1979
                                                                                                                1981
                                                                                                                1983
                                                                                                                1985
                                                                                                                1987
                                                                                                                1989
                                                                                                                1991
                                                                                                                1993
                                                                                                                1995
                                                                                                                1997
                                                                                                                1999
                                                                                                                2001
                                                                                                                2003
                                                                                                                2005
                                                                                                                2007
                                                                                                                2009
                                                                                                                2011
                  2011     2012          2013 (to Jul)                                                           Rice      Wheat
Source: Authors, using CSO data.                                                 Source: Authors, using DES, MOA data.

Nevertheless, rising prices for food grains have                                 proportion of whom are landless) and of
been annoyingly persistent (2010 and 2011 are                                    small-scale farmers (who are net buyers of food)
exceptions). After 2011, food grains reemerged                                   (Deaton 1989; Ivanic and Martin 2008, Ivanic,
as the largest single contributor to the food                                    Martin and Zaman 2011, Anderson, Ivanic and
price rise, driven by the prices of cereals and                                  Martin 2014).22
the sheer size of cereals’ continuing large
share in the consumption basket. Real prices of                                  Estimates of nutrient demand elasticities
staples have been rising since the early 1990s.                                  reaffirm the intuitive conviction that higher
This trend seems to follow that of the real                                      prices affect the nutrients consumed, possibly
Minimum Support Prices (MSPs) for rice and                                       contributing to malnutrition as households
wheat (Figure 5), which have been empirically                                    (rural and urban) reduce calorie and protein
shown to drive domestic prices for these cereals                                 consumption through own and cross-price
(Bathla 2011; Acharya et al. 2012).                                              effects (Gaiha et al. 2012).23 Given that cereals
                                                                                 contribute 64 and 54 percent of calories (Gaiha
Farmers (the net-sellers) benefit from rising                                    et al. 2012, using 2009/10 NSS data), an increase
real prices, which translate directly to higher                                  in the cereal price may reduce food intake as
incomes. Food price increases driven by                                          households struggle to meet expenses other
support prices may be masking underlying                                         than food (Deaton and Dreze 2009). National
inefficiencies, however, as sluggish growth                                      Sample Survey (NSS) data show that households
in productivity (yields) may be driving up the                                   are spending less of their incremental income
costs of cultivation (cost per quintal), which
                                                                                 22	 Using 2003-04 data, Rawal (2008) estimated 42 percent of
in turn provide the rationale for pushing up                                         rural households to be landless (that is, they own no land
MSPs (Rao and Dev 2010). In any case, rising                                         other than the homestead).
                                                                                 23	 Gaiha et al. (2012), using NSS data, find price effects to
prices can have significant consequences,                                            have weakened over time from 1993-94 to 2009-10, but the
particularly on the welfare of the poor (a large                                     negative elasticities remain significant.




                                                                                                                                            9
  Republic of India: Accelerating Agricultural Productivity Growth




             elationship between domestic and external wheat prices in surplus producing areas, 2005
  Figure 6: R
            Rs./MT (left panel) and distortion in agricultural incentives (right panel)

                              16                                                    0.6

                              14
Real 2005 Thousnd Rupees/MT




                                                                                    0.4

                              12                                                    0.2
                              10
                                                                                      0
                              8
                                                                                    -0.2
                              6
                                                                                    -0.4
                              4
                                                                                    -0.6
                              2
                                                                                    -0.8
                              0



                                                                                           1965
                                                                                           1967
                                                                                           1969
                                                                                           1971
                                                                                           1973
                                                                                           1975
                                                                                           1977
                                                                                           1979
                                                                                           1981
                                                                                           1983
                                                                                           1985
                                                                                           1987
                                                                                           1989
                                                                                           1991
                                                                                           1993
                                                                                           1995
                                                                                           1997
                                                                                           1999
                                                                                           2001
                                                                                           2003
                                                                                           2005
                                                                                           2007
                                                                                           2009
                                   1990
                                   1991
                                   1992
                                   1993
                                   1994
                                   1995
                                   1996
                                   1997
                                   1998
                                   1999
                                   2000
                                   2001
                                   2002
                                   2003
                                   2004
                                   2005
                                   2006
                                   2007
                                   2008
                                   2009
                                   2010
                                   2011
                                   2012



                                   Domestic Price   Import Parity   Export Parity                            NRA Agric.      RRA

  Source: Authors, using Anderson and Nelgen (2012) database.
  Note: RRA = Relative Rate of Assistance; NRA = Nominal Rate of Assistance.



  on food and more on durable goods and services                                    Trade Policy and Cereal Prices
  (health and education).
                                                                                    India’s trade policies have been unstable
  It is understandable that real prices of food                                     but restrictive (Gulati, Jain, and Hoda 2013),
  other than cereals would rise with rising                                         so export-led prices have not been a factor
  incomes and demand, but the recent sharp                                          in the rise in food prices.24 The government
  upward trend in cereal prices is puzzling,                                        has periodically banned trade in staples (rice
  especially because per capita cereal                                              and wheat) other than aromatic rice, most
  consumption is falling, cereal production is                                      recently between 2008 and 2011. The extent
  growing steadily (notwithstanding the annual                                      of protection is demonstrated by the limited
  fluctuations caused by adverse weather), and                                      transmission of global price volatility to Indian
  food stocks remain at all-time highs. Analysts                                    markets (Imai, Gaiha, and Thapa 2008; Rajmal
  have raised questions about the irony of                                          and Misra 2009; Dasgupta, Dubey, and Satish
  amassing large food grain stocks when prices                                      2011). Gouel, Gautam and Martin 2014) find that
  of staples are high (Chand et al. 2010). Two                                      Indian trade policies offset 94 percent of global
  sources of “leakage” from a purely domestic                                       price volatility for wheat, confirming the great
  supply-demand regime could influence prices:                                      insulation of Indian cereal markets. Figure 6
  trade (through imports or exports, if external                                    (left panel) shows the stability in domestic
  prices are higher than domestic prices) and                                       wheat prices relative to global prices.
  stocks (through excessive diversion of grain
                                                                                    24	 On the import side, oilseeds and pulses are major imports,
  into silos or ill-timed procurement/release
                                                                                        which are likely transmitting global prices to domestic prices
  operations).                                                                          for these commodities.




                              10
                                 Chapter 2: The Context and Conundrums of Contemporary Agriculture




Despite trade restrictions, Indian prices have      this study, but it is the subject of a separate
tracked the long-term trend in global prices, but   ongoing technical study (Gouel, Gautam and
without the short-term volatility (GOI 2013b).      Martin 2014).
India has also generally maintained domestic
prices between import and export parity             More good news for farmers comes with the
prices, making commercial trade economically        shift away from a bias against agriculture
unviable in most years (Figure 6). This             (implicit tax) in domestic incentives toward
alignment of domestic and external prices has       a more supportive regime, as indicated by
been achieved as a managed policy rather than       the trend in the Relative Rate of Assistance to
a market- or trade-determined outcome, with         agriculture (Figure 6, right panel) (Pursell, Gulati
the government using the global price trend as      and Gupta 2009, Anderson and Nelgen 2012).
a key determinant of MSPs (Acharya et al. 2012).    The implicit taxation of agriculture observed
                                                    in the 1970s–90s has steadily improved to
As noted, farmers benefit from cereal prices        a more neutral and often more favorable
that appear globally competitive but do not         (subsidy) regime in the 2000s. Other than the
experience the volatility of global prices. The     explicit output price policy, the farmers are
two policies—trade and MSP-driven storage           also benefiting more significantly through
(discussed later)—are inconsistent, however,        input subsidies and other forms of support, as
in the sense that the restrictive trade policy      reflected in the positive and rising Nominal
pushes prices down (favoring consumers), while      Rate Assistance (NRA) for agriculture since the
the MSP-based storage policy effectively creates    late 1990s. The gross subsidy equivalent of this
a floor price (favoring farmers). Despite prices    assistance amounts to 30–40 percent of the total
being competitive, produce cannot find its          value of agricultural production, indicating
way out to external markets, so national cereal     the significance of subsidies to agriculture
stocks are ballooning, because the government       apart from the border distortions (which seem
is obligated to buy whatever farmers have to        relatively minor) indicated by the difference in
offer—at least in the traditional, politically      domestic and global prices.
influential grain-basket states. As might
be expected, bulging food stocks entail a
                                                    Food Grain Management and
significant budgetary cost for the government
(GOI 2013a). This experience is not unique to
                                                    Food Prices
India; the United States, European Union, and       In principle, stock operations should be an
Australia have faced identical situations. In       infra-marginal transfer; as such, they should not
those instances, governments with open-ended        be a cause for food price increase. In practice in
public procurement systems and large food           India, however, stock-management appears to
stocks subsequently either abandoned their          be suboptimal (Basu 2011).25
programs or adopted alternative mechanisms
with significant added costs to reduce the          25	 The government is obligated to buy at the MSP. Procurement
                                                        remains biased in favor of the traditional “grain baskets” in
pressure to buy (such as acreage set-aside              the country, but there is no defined, rules-based “release”
schemes). Finding viable and more cost-effective        policy, effectively making it very inefficient and cumbersome
                                                        to release stocks when necessary—whether to release
options for managing food grains is clearly a           excessive stocks or to moderate rising prices, as the case
priority. Such analysis is beyond the scope of          may be.




                                                                                                               11
Republic of India: Accelerating Agricultural Productivity Growth




Figure 7: Cereal trade, procurement, and food price index, 1988–2012
                                                  20                                                                                                                                                                      0.4
    Net Procurement; Net Exports (Million Tons)




                                                  15                                                                                                                                                                      0.3




                                                                                                                                                                                                                                 Relative Food Price Rise (%)
                                                  10                                                                                                                                                                      0.2


                                                    5                                                                                                                                                                     0.1


                                                    0                                                                                                                                                                     0
                                                        1990

                                                               1991

                                                                      1992

                                                                             1993

                                                                                    1994

                                                                                           1995

                                                                                                  1996

                                                                                                         1997

                                                                                                                1998

                                                                                                                       1999

                                                                                                                              2000

                                                                                                                                     2001

                                                                                                                                             2002

                                                                                                                                                    2003

                                                                                                                                                           2004

                                                                                                                                                                  2005

                                                                                                                                                                         2006

                                                                                                                                                                                2007

                                                                                                                                                                                       2008

                                                                                                                                                                                              2009

                                                                                                                                                                                                     2010

                                                                                                                                                                                                            2011

                                                                                                                                                                                                                   2012
                                                   -5                                                                                                                                                                     -0.1


                                                  -10                                                                                                                                                                     -0.2

                                                                      Net Procurement                     Net Exports                        Food Prices Relative to All Commodities (%)
                                                  -15                                                                                                                                                                     -0.3
Source: Authors, using DES, MOA, and UN COMTRADE data.


Figure 7 depicts net exports (exports less                                                                                                  The impact of public grain management
imports) of the main cereals (rice and wheat)                                                                                               operations can also be gauged by the sheer
and net procurement (procurement less                                                                                                       magnitude of the current level of stocks and the
offtake by various programs such as the Public                                                                                              share of procurement in total production. In
Distribution System). The figure also shows                                                                                                 2013, stocks peaked (in June) at over 77 million
the extent to which food prices (Food Articles                                                                                              tons, almost two and half times the highest
Index) deviate from the overall Wholesale Price                                                                                             established norm of 32 million tons (for
(All Commodities) Index. Net exports and food                                                                                               midyear). As a share of production, government
prices show no obvious correlation, but net                                                                                                 procurement has increased steadily over time
procurement and price changes increasingly                                                                                                  (Annex 2, Figure A.2.3). Government procurement
appear to coincide beginning in the early 1990s.                                                                                            now represents 32 percent of the rice and
Simple statistical tests confirm that after 1995,                                                                                           46 percent of the wheat produced in the country,
stock operations are correlated with food prices                                                                                            which amounts to 42 percent of the marketed
(the association with exports is not statistically                                                                                          surplus for rice and 66 percent of the marketed
significant), suggesting that procurement in                                                                                                surplus for wheat, effectively dominating and
excess of releases and sales is diverting grain                                                                                             influencing these markets. Given the scale of
from the market to already overflowing silos.26                                                                                             intervention, MSP has effectively become the
                                                                                                                                            floor price in the major producing areas, with
26	 Other than the large exports in 2012, when the most recent                                                                              the rising real MSP pushing up the cost of basic
    ban was lifted, notable exports were in the early 2000s when                                                                            staples and adding to upward price pressures.
    the previous ban (1996–2000) was lifted. Those exports were
    triggered by subsidized stock releases (Below Poverty Line
    prices) for direct exports (Ganesh-Kumar et al. 2010). The
    situation in the early 2000s was also an outcome of excessive
    stock accumulation during the previous ban of 1996–2000.
                                                                                                                                            Rising Real Rural Wages
    Net procurement, on the other hand, was lowest in 2002/03
    owing to the severe drought at the time (though stocks
                                                                                                                                            From a poverty perspective, the good news (at
    continued to be released for exports).                                                                                                  least in the short run) is that the immediate


   12
                                            Chapter 2: The Context and Conundrums of Contemporary Agriculture




negative consequences of higher food prices                         (for exports and the rapidly emerging domestic
have been compensated by rising real wages.                         poultry industry), and soybeans (for exports of
Jacoby (2013) finds that nominal wages for                          soy de-oiled cake as animal feed).
manual labor (within and outside) agriculture
have responded elastically to higher food prices,                   These developments aside, the overall production
with wages rising faster in rural districts that                    structure of the sector has changed relatively
saw greater increases in crop prices between                        slowly (Annex 2, Figure A.2.4), with the main
2004 and 2009. Wage adjustments helped to                           cereals (rice and wheat) continuing to dominate
protect the welfare of the poor, apparently more                    the area allocation. The slow transformation,
effectively than the Public Distribution System.                    even in the face of strong growth in demand
Gulati, Jain, and Saltija (2013) find that real                     (for example, for horticultural commodities),
wages have risen over the long term, growing at                     reflects the incentives facing farmers. The price
a consistent and more rapid pace in the 1990s                       policy, specifically the MSPs for rice and wheat,
than in the 2000s, despite the much-debated                         has been highly successful, inducing a supply
impact of the Mahatma Gandhi National Rural                         response for cereals by keeping prices high,
Employment Guarantee Act (MNREGA). After                            while trade and storage policies have been
falling in the early part of the decade since                       equally successful in stabilizing prices (at least in
2000, real wages rose sharply after 2006/07 and                     the major grain-basket areas for these crops). The
resumed their long-term upward trend. The                           resulting low risk-return ratio for cereals creates
study also finds that while MNREGA does put                         strong incentives for farmers to emphasize
upward pressure on wages, the “pull” factors                        cereals in their crop portfolios—as reflected in
associated with growth (GDP, agricultural GDP,                      area allocation—rather than venturing into crops
or construction-GDP) are much more significant                      that offer relatively higher returns but also face
factors driving wages.27                                            relatively greater market price risks.

                                                                    Rising real prices are good news for farmers
Persistent Questions about the                                      (who are net-sellers of food) and rising real
Sustainability of Growth                                            wages for those engaged in wage labor (on and
                                                                    off the farm), but both trends have implications
The discussion above suggests that the
                                                                    for the sustainability of the ensuing growth.
constraints to growth appear to be less binding
                                                                    The critical question is whether the changed
on the demand side. Consumer incomes and
                                                                    incentives afforded by rising real output prices
habits appear to be driving the demand for high-
                                                                    are accompanied by growth in agricultural
value agricultural commodities and, ironically,
                                                                    productivity. Rising real wages have potential
public procurement maintains a steady source
                                                                    implications for employment both on and
of absorption (even as consumer demand is
                                                                    off the farm, as they are likely to induce
shrinking) for cereals. A combination of domestic
                                                                    greater mechanization on the farm and more
industrial demand as well as export demand has
                                                                    capital intensity off the farm. Understanding
driven specific “green shoots” in the economy,
                                                                    developments in the underlying productivity
such as cotton (for exports and textiles), maize
                                                                    (total as well as factor) is important to assess
                                                                    whether they are underpinned by a sustainable,
27	 Gulati, Jain, and Saltija (2013) find that MNREGA does have a
    significant impact on wages, but the broader growth “pull”
                                                                    virtuous circle of growth—as the next chapters
    factors are 4–6 times more effective in raising incomes.        will attempt to discern.


                                                                                                                    13
                     Temporal Trends in
3                    Agricultural Output and
                     Productivity



Evolution of Agricultural Growth                             usual—is more complicated. Indian agriculture
                                                             has seen a series of significant transformations.
Large annual fluctuations in agricultural                    The green revolution (in cereal production)
growth in India are a direct outcome of variable             was succeeded by the white revolution (in milk
rainfall, as over 60 percent of cultivated area is           production) and the more recent “rainbow”
still rainfed (Figure 8). The long-term average              revolution (in high-value production of fruits
growth of just under 3.0 percent has improved                and vegetables) as the demand patterns shifted
almost imperceptibly over the past six decades               with rising incomes in the rest of the economy.
(see the linear trend in the figure), although               Agriculture also felt the effects of the broader
the reduced dispersion in recent years suggests              economic reforms of the early 1990s and the
increased stability.                                         agricultural policy reforms of the mid-1990s.
                                                             The outcomes of those reforms continue to be
This rather bleak picture raises questions about             debated.28
the prospects for agricultural growth to increase
shared prosperity, yet the underlying story—as               More nuanced analyses of temporal and spatial
                                                             patterns of growth illuminate the major factors
Figure 8: Agricultural Growth Rate, 1952–2012                behind agriculture’s varied performance and
20                                                           help to identify the appropriate policy levers.
                                                             Chand and Parappurathu (2012) use decadal
15
                                                             trend growth rates to analyze performance
10
                                                             28	 Agricultural trade liberalization has been debated for a
 5                                                               long time in India (see Chand and Bajar 2012). One school
                                                                 of thought (the so-called “trade liberalization proponents”)
 0                                                               views trade restrictions as a major constraint to growth and
                                                                 diversification, associated with a significant opportunity cost
                                                                 in terms of potential export earnings (for example, see Gulati
 -5
                                                                 et al. 2012; Ganesh-Kumar and Parikh 1998).Trade is also
                                                                 viewed as a better option for ensuring domestic supply and
-10                                                              price stability (Jha and Srinivasan 1999). The other school of
                                                                 thought is more skeptical of gains from trade, arguing that
-15                                                              international prices are highly distorted, do not represent the
      1952
      1955
      1958
      1961
      1964
      1967
      1970
      1973
      1976
      1979
      1982
      1985
      1988
      1991
      1994
      1997
      2000
      2003
      2006
      2009
      2012




                                                                 opportunity cost of resources, and are much more volatile
-20                                                              than may be desirable. Hence the proponents of this view
        Annual Growth Rate     Linear (Annual Growth Rate)       argue for “strategic openness” rather than full liberalization
                                                                 and promote reliance on buffer stocks rather than trade to
Source: Authors, using CSO data.                                 ensure stable prices (Chand 2003a, Chand and Bajar 2012).




                                                                                                                         15
Republic of India: Accelerating Agricultural Productivity Growth




           emporal performance of Indian agriculture GDP: Trend growth rate and growth volatility for
Figure 9: T
          decades ending 1960/61 to 2012/13
    4.0%                                                                                                                                                                                                10%

                                                                                                                                                                                                        9%
    3.5%
                                                                                                                                                                                                        8%
    3.0%
                                                                                                                                                                                                        7%

    2.5%
                                                                                                                                                                                                        6%

    2.0%                                                                                                                                                                                                5%

                                                                                                                                                                                                        4%
    1.5%
                                                                                                                                                                                                        3%
    1.0%
                                                                                                                                                                                                        2%
    0.5%
                                                                                                                                                                                                        1%

    0.0%                                                                                                                                                                                                0%
           1960
                  1962
                         1964
                                1966
                                       1968
                                              1970
                                                     1972
                                                            1974
                                                                   1976
                                                                          1978
                                                                                 1980
                                                                                        1982
                                                                                               1984
                                                                                                      1986
                                                                                                             1988
                                                                                                                    1990
                                                                                                                           1992
                                                                                                                                  1994
                                                                                                                                         1996
                                                                                                                                                1998
                                                                                                                                                       2000
                                                                                                                                                              2002
                                                                                                                                                                     2004
                                                                                                                                                                            2006
                                                                                                                                                                                   2008
                                                                                                                                                                                          2010
                                                                                                                                                                                                 2012
                                                 Trend Growth Rate (Left Axis)                                              Growth Volatility (Right Axis)

Source: 
        Authors using CSO data, following Chand and Parappurathu’s (2012) approach for trend growth rates.


over time. Figure 9 shows how growth evolved                                                             Based on statistically estimated endogenous
since 1950. The solid line represents the trend                                                          break points, Chand and Parappurathu (2012)
growth rates in agricultural GDP for each of                                                             identify six distinct periods of performance
the 10-year periods ending in 1960/61–2012/13.                                                           (Table 1). After the initial decline into the 1960s,
The dashed line indicates the change in                                                                  which led to the concerted policy push for
volatility of growth (standard deviation of                                                              the green revolution, the late 1960s to early
growth for the same 10-year periods). When                                                               1970s saw the initial green revolution results.
the noise associated with annual fluctuations                                                            Sustained growth followed through the 1980s
in production is filtered out, distinct episodes                                                         as the green revolution technology spread to
of varying performance become apparent.                                                                  other parts of the country. The early 1990s were
Volatility increased until 1980 and then declined                                                        a period of diversification, with growth rates
markedly. Growth remained stubbornly below                                                               reaching their highest levels of just over 3.5
the 4 percent target throughout but reached                                                              percent. A prolonged slowdown followed from
its highest level in the most recent period.                                                             about 1996/97 to 2004/05. Growth remained
The high growth rate of about 7.9 in 2010/11                                                             positive throughout but slowed progressively to
followed by 3.9 percent in 2011/12 brought the                                                           a low of 2 percent.
trend rate to the high of 3.6 percent reached in
the mid-1990s, but slower growth of 1.9 percent                                                          This deceleration has been cited as an
for 2012/13 has kept the target out of reach.                                                            outcome of the general neglect of agriculture



  16
                                             Chapter 3: Temporal Trends in Agricultural Output and Productivity




Table 1: Sectoral Trend Growth Rates by Period (%/yr)

                                       PGR             EGR           WTD           DIV              PR            REC
                                    (1960/61–       (1968/69–      (1975/76–    (1988/89–       (1995/96–      (2004/05–
Sector                               1968/69)        1975/76        1988/89)     1995/96)        2004/05)       2010/11)
Agric. and Allied Activities           1.03            1.98           2.42         3.24            2.35           3.31
Agriculture*                           0.70            1.93           2.71         3.21            2.30           3.37
  Crops (VOP)                          1.11            1.90           2.56         2.64            1.88           3.01
  Livestock (VOP)                      0.40            2.69           4.89         4.12            3.43           4.29
Forestry and Logging                   3.70            2.01          -1.77         0.74            2.05           2.25
Fishery                                3.91            4.19           3.45         7.37            3.28           4.42
Nonagriculture                         4.90            3.67           5.23         5.91            7.05           9.68
All sectors                            3.19            2.99           4.25         5.14            5.95           8.57
Sources: Chand and Parappurathu 2012.
Notes: The original table by the authors is based on data to 2010/11. The growth rate given in the text is slightly
higher, as it is based on data to 2011/12. PGR = Pre-green revolution period (1960/61–1968/69); EGR = early
green revolution period (1968/69–1975/76); WTD = period of wider technology dissemination (1975/76–1988/89);
DIV = period of diversification (1988/89–1995/96); PR = post-reform period (1995/96–2004/05); and REC = period of
recovery (2004/05–2010/11).


that occurred as policy shifted attention                         the last (recovery) period appears equally
away from agriculture in favor of the faster-                     widespread, with historically high growth rates
growing secondary and tertiary sectors.                           in fruits, vegetables, fiber crops (primarily
Because the deceleration coincided with the                       cotton), and pulses. At the same time, overall
general economic reforms of the 1990s and                         growth in the value of output of the crop
the agricultural reforms initiated in earnest                     sector remains modest at about 3 percent.
around 1994, the reforms are also generally                       The value of output in the livestock sector has
perceived to have had a significant adverse                       grown significantly faster than crops through
impact on the sector (reflected in also Chand                     all phases since the 1970s. Although it too
and Parappurathu’s classification of these years                  experienced a slowdown between 1995/96 and
as the “post-reform period”). The most recent                     2004/05, it fared much better.
period (2004/05–2011/12) shows a marked return
to growth of about 3.5 percent per annum.                         A number of explanations are forwarded in
                                                                  the literature for the performance of Indian
Subsector and crop-specific growth rates                          agriculture, especially the deceleration after
follow these broad trends, with some notable                      almost two decades of sustained expansion
exceptions, indicating that the slowdown in                       (Chand, Raju, and Pande 2007; GOI 2007). But
the post-reform period was widespread (Chand                      two sets of fundamental questions remain
and Parappurathu 2012).29 The rebound during                      (and were raised by policy makers during
                                                                  consultations for this study) about the causes of
29	 Bhalla and Singh (2001, 2010) provided earlier accounts of    the slowdown and the post-2004/05 recovery:
    the deceleration. Notable exceptions to the general trends
    included high growth rates of soybeans in the 1980s and
                                                                     1.	 What were the main factors behind the
    1990s, rapid growth in maize during the period of slowdown,
    and the very high growth rates for cotton after 2004/05.             slowdown? Was it the reforms of the



                                                                                                                      17
Republic of India: Accelerating Agricultural Productivity Growth




       1990s, contrary to what proponents of          the long run, their immediate role in the post-
       reform expected? Is the post-2004/05           reform slowdown is less obvious. They do not
       rebound just a result of good weather,         explain the seemingly discrete downward
       or is it a robust and sustainable move         shift in the trend growth rate in agriculture
       toward a different growth trajectory?          starting in 1997, or what seems to be the
                                                      equally decisive turnaround in 2005 (both are
   2.	 Digging deeper into the drivers of
                                                      visually obvious, without any formal statistical
       growth, how has technology driven
                                                      tests). For example, private investment in
       performance in agriculture? Has the
                                                      groundwater has been the major driver of
       policy framework been supportive of
                                                      irrigation, now accounting for almost two-thirds
       growth, given the vigorous debates on
                                                      of all irrigated area. Capital formation data
       various aspects of agricultural policy?
                                                      show that private investment actually went up
       Most importantly, what are the strategic
                                                      prior to and during the post-reform slowdown
       implications for the sector and for
                                                      (Figure 10). Public investment rose significantly
       India’s broader development strategy?
                                                      after 2003/04 and likely played a role in the
                                                      recovery after 2004/05, but no obvious trigger is
The first set of questions is addressed in the
                                                      apparent from the pre-1997 trend.
remainder of this section; the second set will
be addressed in subsequent sections. Given the
                                                      Similarly, despite rising wages and farmer’s
limitations of specific analytical tools and the
                                                      perceptions of increasing costs of cultivation
data available, it is important to highlight that
                                                      (Rao and Dev 2010), agricultural terms of
the analysis presented here does not provide
                                                      trade provide no explanation for the sudden
definitive answers, but rather adds new and
                                                      downturn in 1997 (Figure 11). Terms of trade
rigorous findings to promote discussion and
                                                      improved for agriculture from the mid-1980s
perhaps stimulate further investigation to
better inform policy and development debates.
                                                                Gross fixed capital formation in
                                                     Figure 10: 
                                                                agriculture
Why Did Growth Stagnate from                                                        140
                                                    Rs. Thousand Crores (2004/05)




1997 to 2005?                                                                       120

Several explanations have been forwarded                                            100
for the growth slowdown between 1997 and
                                                                                     80
2005. They include slow generation of new
technologies, poor dissemination of existing                                         60

ones, weak and inefficient institutions, poor                                        40
governance, and perhaps most critically the
                                                                                     20
inadequate investment in public goods, as the
fiscal space was crowded out by the provision of                                      0
                                                                                          1993-94
                                                                                          1994-95
                                                                                          1995-96
                                                                                          1996-97
                                                                                          1997-98
                                                                                          1998-99
                                                                                          1999-00
                                                                                          2000-01
                                                                                          2001-02
                                                                                          2002-03
                                                                                          2003-04
                                                                                          2004-05
                                                                                          2005-06
                                                                                          2006-07
                                                                                          2007-08
                                                                                          2008-09
                                                                                          2009-10
                                                                                          2010-11




private goods.

While these determinants of agricultural                                                  Public Sector   Private Sector   Total

productivity and growth are important over           Source: Authors, using CSO data.




  18
                                               Chapter 3: Temporal Trends in Agricultural Output and Productivity




Figure 11: Agriculture terms of trade                             growth after the reforms were introduced. The
140                                                               possible roles of rainfall and policy reforms
130                                                               in the slowdown are explored in greater
120                                                               detail next.
110
100
                                                                  Rainfall shocks and productivity
 90
                                                                  growth
 80
 70                                                               To what extent can rainfall shocks explain
 60                                                               changes in productivity growth, especially
      1960
      1962
      1964
      1966
      1968
      1970
      1972
      1974
      1976
      1978
      1980
      1982
      1984
      1986
      1988
      1990
      1992
      1994
      1996
      1998
      2000
      2002
      2004
      2006
      2008
      2010
      2012
                                                                  the deceleration of 1997–2005? The answer
                                                                  to this question is important because it has
Source: Authors, using RBI online database.
                                                                  implications for policy and for strategic
Note: Derived as Ag. All. Act. GDP deflator/Total GDP Deflator.
                                                                  tradeoffs, for example between investments
                                                                  to mitigate sustained rainfall shocks (as may
to about 2000 and seemed to have followed the                     be expected with climate change) versus
slowdown rather than precede it.                                  investments to adapt to large but idiosyncratic
                                                                  shocks.
Two factors that may help understand the
growth deceleration are rainfall and policy                       For example, investments to mitigate sustained
reforms. Virtually all analyses of Indian                         shocks would need to focus on safeguarding
agriculture recognize the importance of                           and economizing the use of scarce “water
rainfall for agricultural performance, but                        stocks” to deal with sequential droughts,
temporal or cyclical rainfall patterns have not
                                                                  calling for more sustainable irrigation options
been sufficiently analyzed for their potential
                                                                  (such as more efficient irrigation techniques,
to explain growth patterns in agriculture.
                                                                  policies to promote water use efficiency, water
And although the performance of agriculture
                                                                  harvesting, and so forth), developing better
following the reforms of the 1990s continues
                                                                  technological solutions (such as crop varieties
to be debated,30 Kumar and Jain (2012) doubt
                                                                  for rainfed ecologies and the diversification
that policy reforms can explain the slowdown,
                                                                  of crops and farming systems), as well as
considering that growth has rebounded after
                                                                  diversifying and transforming sources of
2004/05 with no major policy change. It is
                                                                  livelihoods. Adapting to large but idiosyncratic
plausible that the agricultural sector went
                                                                  shocks would involve better ex-post coping
through an adjustment when the reforms
                                                                  mechanisms, including efficient markets to
were first introduced, or that the agricultural
                                                                  ensure smoother flows of food to affected areas,
programs initiated during the Eleventh Five
                                                                  safety net programs, and smarter agricultural
Year Plan (around 2004/05) have contributed to
                                                                  insurance mechanisms. In reality, both sets
the subsequent growth. It is thus important to
                                                                  of options are necessary, but it is important
explore which factors were most likely to have
contributed to the slowdown in agricultural                       strike an informed strategic balance between
                                                                  the two, especially given the increasing need to
30	 See Chand and Parappurathu (2012); Gulati et al. (2013);
                                                                  deal with changing and unpredictable climatic
    Chand and Bajar (2012).                                       cycles and events.



                                                                                                              19
Republic of India: Accelerating Agricultural Productivity Growth




A district-level picture of productivity                                sustained decline in growth rates between 1997
and rainfall between 1970 and 2007                                      and 2004 was unusual.34 As expected, rainfall
                                                                        anomalies explain the deviations from trend
Given the heterogeneity of India’s agroecological                       in agricultural production and productivity.35
environments and especially the variability in                          Figure 12 depicts the correlation between large
rainfall across those environments, disaggregated                       negative rainfall shocks and the prolonged
data can provide a more credible analysis and                           stagnation in productivity growth between
better insights than national averages or and                           1999 and 2004; it also shows the correlation
trends. Kshirsagar and Gautam (2013) use district                       between surges in productivity and favorable
data on production and rainfall to develop a                            rainfall in 1988–90 and 1993–97.36 Aside from
picture of aggregate agricultural productivity.31                       influencing productivity, rainfall has an even
Their analysis finds a reasonably convincing                            larger influence on deviations in total production
association between growth and deviations in                            as households adjust the area they plant in
rainfall from the long-term rainfall trend, which                       response to rainfall deviations. These impacts
provides additional insights into the observed                          on productivity also help explain the decline
growth episodes.                                                        in real wages between 1999 and 2004 noted by
                                                                        Lanjouw and Murgai (2008) and Gulati, Jain, and
Agricultural productivity—measured here                                 Saltija (2013).
as the real value of agricultural production
per hectare—grew by 2.5 percent on average
between 1970 and 1996. The value of                                     The varied incidence and impact of
production and productivity between 1997                                sustained rainfall shocks between 1970
and 2004 declined in parallel with GDP. As                              and 2007
discussed, agricultural productivity began                              Cereal crops are better able to withstand
to recover after 2004 at a steady but slow                              rainfall shocks than noncereal crops,37 so risk-
1.9 percent up to 2007.32 All crops showed                              averse households often respond by favoring
significant fluctuations in productivity around                         cereals in their crop portfolios (Annex 3,
trend, but the productivity of crops other than                         Figure A.3.4). Cereals may be the less risky
cereals proved more volatile.
                                                                        34	 Kshirsagar and Gautam’s analysis used data from 1970
The fluctuations in productivity occurring                                  onward. Extending the analysis back into the 1960s shows
around 2000 were not unprecedented,33 but the                               the relationship between the severe rainfall shock of the
                                                                            1960s and the widespread famines experienced then.
                                                                        35	 Anomalies are defined as percentage deviations from the
31	 The data are from the district database compiled by ICRISAT             period average.
    and NCAP. Data for consistent dynamic analysis are available        36	 The data used in Figure 12 are for the 207 districts that
    for 1970–2007 and for 297 “original” rural districts (districts         have complete data on annual rainfall levels, area and
    existing in 1970). Subsequent divisions of districts and states         production in the ICRISAT/NCAP district database. The
    are mapped to these original administrative units. These data           estimated district rainfall anomalies bear a very close
    cover about 88 percent of the aggregate cereal production,              relationship with anomalies estimated using the national-
    providing a reasonably representative picture at the national           level aggregate monsoon rainfall data from the Indian
    level. See ICRISAT (2013) for a detailed description of the data.       Meteorological Department, as shown in Annex 3, Figure
32	 The data are relatively more consistent and reliable for                A.3.1. The deviations in production and productivity using all
    cereals than for noncereals. Cereal productivity increased at           297 districts also show a very similar association with rainfall
    3 percent per year between 1970 and 1996, and at 2.5 percent            anomalies from the national aggregate monsoon rainfall data
    over the longer period from 1970 to 2007.                               from IMD, shown in Annex 3, Figure A.3.2.
33	 Deep declines occurred in the early 1970s and in 1979, and          37	 Figure A.3.3 in Annex 3 shows relatively larger deviations in
    less severe declines occurred in the late 1980s.                        the productivity of noncereal crops.




   20
                                                           Chapter 3: Temporal Trends in Agricultural Output and Productivity




Figure 12: Relationship between agricultural productivity and rainfall

                           20



                           10
   Percent Deviation (%)




                            0



                           -10



                           -20



                           -30

                                 1970               1980                   1990                         2000                          2010
                                                                           Year
                                        Aggregate Rainfall Anomaly (%)                       Trend-Deviation in Real Ag. Production
                                        Trend-Deviation in Cropped Area                      Trend-Deviation in Real Ag. Productivity

Source: Kshirsagar and Gautam 2013.


choice, but they are less remunerative than                                districts.38 One explanation is that rainfall
noncereal crops, so producers’ incomes grow                                shocks are idiosyncratic. Another may be that
more slowly.                                                               irrigation in some districts compensates for
                                                                           rainfall deficits. Over the long run, evidence
The consequences of these rainfall shocks                                  shows that districts in the Semi-Arid Temperate
are thus significant—in total production as                                Zone are generally less vulnerable to rainfall
well as in the relative differences between                                shocks.39 Even so, the districts that experienced
cereal and noncereal crops. District data after                            the most severe rainfall shocks between
2007 are not available, but rainfall anomalies                             2000 and 2007 were not those with greater
from the longer rainfall time series (discussed                            vulnerability to rainfall shocks over the long
later) show that rainfall was below average in                             term, namely districts in the Arid or Semi-Arid
2008–12, with 2009 being an exceptionally dry
year. Rainfall risk may be partly contributing                             38	 Kshirsagar and Gautam (2013) estimate rainfall sensitivity
                                                                               using dynamic regressions for districts with a minimum of
to the sluggish supply response in noncereal                                   25 years of observations—277 districts (out of a total of 297).
crops, amplifying the upward pressure on                                       Of those 277 districts, 121 emerged as “rainfall-dependent,”
                                                                               confirming that districts are not equally affected by
their prices.                                                                  rainfall shocks.
                                                                           39	 The zones are grouped into the four broad agro-ecological
                                                                               classifications developed by ICRISAT. Most districts in Uttar
There is considerable variation in the                                         Pradesh, Punjab, and Haryana are grouped with the Semi-
sensitivity of productivity to rainfall across                                 Arid Temperate Zone.




                                                                                                                                        21
Republic of India: Accelerating Agricultural Productivity Growth




            istribution of rainfall shocks across
Figure 13: D                                                    the post-reform period, sustained rainfall
           zones, 2000–07                                       scarcity depleted the water available for canals,
             Humid
                                                                tanks, and dug wells, slowing their importance
Semi-Arid Temperate                                             as sources of irrigation. The reliance on
  Semi-Arid Tropical                                            groundwater, via tubewells, was a relatively
               Arid                                             more important source of irrigation expansion
                       -50          0                   50
                                                                in this period. Over time, however, this growth
                             Rainfall Anomaly (%)               slowed, probably as the potential to install new
                                                                tubewells diminished or existing tubewells
Source: Kshirsagar and Gautam 2013.
                                                                went out of commission as groundwater was
                                                                depleted. The only source of irrigation that
Tropical Zones (Figure 13). Semi-arid temperate                 shows a spike in growth is “other sources”—
districts, which are relatively less vulnerable                 most likely streams and rivers, as in hilly
and more productive on average, were the                        Himachal Pradesh.
worst hit. Given the magnitude and the
succession of negative rainfall anomalies over
                                                                Three questions raised by recent
six years, the cumulative aggregate impact on
agricultural productivity was large.
                                                                rainfall anomalies
                                                                The district analysis of rainfall anomalies
The severity of sustained rainfall deficits is                  between 1970 and 2007 raises three questions.
also evident in patterns of change in irrigation.               First, how do the rainfall anomalies in that
Figure 14 shows the expansion in irrigated                      period compare to rainfall patterns over a
acreage by source from 1970 and 2006. During                    much longer period? Second, are the 1970–2007

Figure 14: Growth in irrigated area by source
12%

10%

 8%

 6%

 4%

 2%

 0%

-2%      NIA_CANAL           NIA_TANK          NIA_TUBE WELL         NIA_WELL          NIA_OTHER     NIA_TOTAL
                                                                                        SOURCE
-4%

-6%

-8%

                                1971-1987           1987-1997    1997-2003      2003-2008

Source: Authors, using ICRISAT-NCAP district database.




   22
                                                                                             Chapter 3: Temporal Trends in Agricultural Output and Productivity




  Figure 15: Historical rainfall patterns, 1901–2010
                                  Distribution of anomalies between and across decades                                                                                       Average anomalies across decades

                            20                                                                                                                          5
Mean Rainfall Anomaly (%)




                                                                                                                           Mean Rainfall Anomaly (%)
                            10

                                                                                                                                                        0
                             0


                            -10
                                                                                                                                                        -5

                            -20


                            -30                                                                                                                        -10
                                   1900s

                                           1910s

                                                   1920s

                                                           1930s

                                                                   1940s

                                                                           1950s

                                                                                   1960s

                                                                                           1970s

                                                                                                   1980s

                                                                                                           1990s

                                                                                                                   2000s




                                                                                                                                                             1900s

                                                                                                                                                                     1910s

                                                                                                                                                                              1920s

                                                                                                                                                                                      1930s

                                                                                                                                                                                              1940s

                                                                                                                                                                                                      1950s

                                                                                                                                                                                                              1960s

                                                                                                                                                                                                                      1970s

                                                                                                                                                                                                                              1980s

                                                                                                                                                                                                                                      1990s

                                                                                                                                                                                                                                              2000s
Source: Calculations based on Monsoon (June–September) national rainfall data from IMD.
Note: The left panel describes the distribution of rainfall anomalies. The white line in the rectangle shows the decade
median. The ends of the rectangles indicate the 25th and 75th percentiles, respectively, and the bars at the end the lines
are the 5th and 95th percentiles. The right panel describes the mean decadal rainfall.


anomalies nationally representative? Finally,                                                                              anomalies (defined as anomalies above +5
what has happened since 2007? Homogenous                                                                                   percent). Instead, the decade had three years
(consistent) rainfall data (totals from June to                                                                            of exceptionally low rainfall: 2004 (fourteenth
September) from the India Meteorological                                                                                   lowest), 2002 (sixth lowest), and 2009 (second
Department (IMD) at the national level, from                                                                               lowest). Finally, the anomalies that explained
1901 to the end of 2010, help to answer those                                                                              the stagnation in agricultural productivity
questions.                                                                                                                 of the early 2000s, analyzed in the previous
                                                                                                                           section, appear to be representative of
The long-term rainfall patterns, over the                                                                                  national patterns.
110 years for which data are available,
do not show a secular monotonic pattern                                                                                    Mechanisms for coping with the risk of
(Figure 15). From the 1900s to 1950s, India
                                                                                                                           rainfall shocks
saw a sustained increase in rainfall, with the
three decades from the 1930s to the 1950s                                                                                  The trends presented here raise important
experiencing above-average rainfall. The 1950s                                                                             policy issues that are beyond the scope of this
ushered in a long-term decline. The most                                                                                   study, but clearly mechanisms to mitigate
unfavorable rainfall patterns—with historic                                                                                the impacts of rainfall shocks as well as to
lows in average rainfall—occurred from 2000                                                                                permit better and faster adjustment will
to 2009. The median anomalies in the 2000s                                                                                 be critical in the future. It will be vital to
are comparable to those in some earlier                                                                                    quantify the costs and benefits from adopting
decades, but unlike other decades, the 2000s                                                                               alternative ex ante risk-mitigation strategies (for
saw no years with substantial positive rainfall                                                                            example, sustainable irrigation, percolation



                                                                                                                                                                                                                                              23
Republic of India: Accelerating Agricultural Productivity Growth




tanks, and other strategies) and ex post risk-                       anomalies in any given year (Figure 16).42 The
sharing strategies (such as regulatory changes                       average sensitivity of districts to a 1 percent
that engender greater trade across districts,                        increase in the rainfall anomaly is 0.82
states, and countries). One advantage of                             percentage points, with considerable variation
risk mitigation policies is that they directly                       across the districts. At the low end (25th
address entitlement risk. In contrast, ex                            percentile), sensitivity is 0.5, while at the high
post risk-sharing strategies rely more on                            end (the 75th percentile) it is 1.2. This suggests
well-functioning markets and/or national                             potential for risk sharing across districts in a
and local governance systems for successful                          given year, which may be more cost-efficient
implementation.40                                                    than (or in addition to) risk sharing across time.
                                                                     Risk sharing across districts is implemented
Given that rainfall anomalies exhibit temporal                       more effectively when local prices reflect local
dependence, and given the growing evidence                           (relative) shortages or surpluses, and it requires
of a relationship between ENSO cycles and                            a smoothly functioning market system.
anomalous monsoon patterns in India, an Early
Warning System could be very effective. Such                         Rainfall shocks and the impact of policy
a system would provide input to extension                            reforms, revisited
services to advise farmers on crop allocation
and input use contingent on expected rainfall.                       The findings above could be refined in several
Consequently, programs that include the                              ways, but the influential role of rainfall shocks
establishment of drought management                                  during the 1997–2004 productivity stagnation
systems and measures to reduce the reliance                          is quite apparent. It is worth noting that
on rainfed agriculture are urgently needed and                       weather shocks have not featured sufficiently
have a larger Net Present Value than may be                          in discussions on the long-term performance
expected.41                                                          of India’s agricultural sector. Given that
                                                                     successive rainfall shocks followed soon after
In addition to ex ante strategies to mitigate                        the policy reforms, it is crucial to revisit the
risk, ex post strategies will remain essential.                      question of the reforms’ impact on agricultural
Rainfall shocks are correlated across districts,                     performance. One approach is to use time-series
but with significant dispersion in rainfall                          data on monsoon rainfall and aggregate data
                                                                     on agricultural GDP to simulate agricultural
                                                                     growth rates that might have prevailed had
40	 Basu (2011) makes a compelling case for the more efficient
    distribution of food grains using direct transfers that
                                                                     rainfall been normal for the entire period
    take advantage of technological advances (smart cards,
    biometric identification). This approach would address food
    security concerns while also (potentially) incentivizing
    agricultural production and improving the functioning of         42	    The standard deviation for the entire sample of
    grain markets.                                                         10,546 annual (district-year) rainfall observations is
41	 A first step in any such cost-benefit analysis—one in which            697.6 millimeters, the between standard deviation (deviation
    the government has invested significant resources (see http://         of the district means from the overall mean) is 598
    www.indiawaterportal.org/taxonomy/3/Aquifer-Mapping)—is                millimeters (297 observations), and the within standard
    to map all aquifers. This information will make it possible to         deviation (deviation of the observations from the district
    quantify the relationship between rainfall and groundwater             mean) is 340.8 millimeters (average of 35.5 observations
    levels under alternative modes of irrigation and farming,              per district). Although some of these “observations” were
    thus making it possible to value (and prioritize) returns to           inferred using interpolation, it is still clear that considerable
    prospective water and irrigation investments.                          rainfall variation is present both within and across districts.




   24
                                                Chapter 3: Temporal Trends in Agricultural Output and Productivity




Figure 16: Distribution of rainfall anomalies across districts and years

    1970
    1971
    1972
    1973
    1974
    1975
    1976
    1977
    1978
    1979
    1980
    1981
    1982
    1983
    1984
    1985
    1986
    1987
    1988
    1989
    1990
    1991
    1992
    1993
    1994
    1995
    1996
    1997
    1998
    1999
    2000
    2001
    2002
    2003
    2004
    2005
    2006
    2007

           -100                      -50                         0                       50                       100
                                                        Rainfall Anomaly (%)
Source: Kshirsagar and Gautam 2013.


(1961 to 2010).43 The actual and counterfactual                      What is remarkable about the trends is the
growth rates for agricultural GDP are averaged                       sharp deviation between the actual and
over rolling 10-year periods, and the decadal                        counterfactual trends starting just after the mid-
averages are plotted in Figure 17.44 The latent                      1990s, following the agricultural sector reforms
long-term growth trends are highlighted using a                      (which in turn followed the general reforms
polynomial trend for each time series.                               implemented in 1991). Assuming normal
                                                                     rainfall, with all else remaining the same, the
                                                                     simulated growth trend shows a significantly
43	 Time-series data on aggregate agricultural GDP from 1950/51
    and monsoon rainfall data from 1900 onward from IMD are          different trajectory than the actual observed
    used to estimate the impact of rainfall on agricultural GDP      growth rate.45 It more dramatically highlights
    growth. Using the estimated parameters, a counterfactual
    scenario of zero rainfall anomaly (normal rainfall) is           the significant impact of the green revolution
    estimated for each year.                                         in the 1960s and 1970s. The counterfactual
44	 The dynamic growth regression explains about 61 percent
    of the variation. Note also that the regression does not
                                                                     scenario further suggests that growth after
    completely eliminate the impact of rainfall but does             the mid-1990s may possibly have been higher
    account for a significant part of it. Furthermore, the general
    equilibrium effects are not accounted for, especially for
    the early years when the Indian economy was much more            45	 Had the anomalies been just random annual events, as in the
    susceptible to rainfall shocks, given the more significant           earlier years, the two trends would have followed each other
    downstream linkages of agriculture with other sectors.               in a parallel fashion.




                                                                                                                               25
Republic of India: Accelerating Agricultural Productivity Growth




Figure 17: Ten-year average agricultural GDP growth rates (actual and counterfactual scenarios)
 6


 5


 4


 3


 2


 1


 0
      1961

             1963

                    1965

                           1967

                                  1969

                                         1971

                                                1973

                                                       1975

                                                              1977

                                                                     1979

                                                                            1981

                                                                                   1983

                                                                                          1985

                                                                                                 1987

                                                                                                        1989

                                                                                                               1991

                                                                                                                      1993

                                                                                                                             1995

                                                                                                                                    1997

                                                                                                                                           1999

                                                                                                                                                  2001

                                                                                                                                                         2003

                                                                                                                                                                2005

                                                                                                                                                                       2007

                                                                                                                                                                              2009
-1
                                                Observed Growth Rate                             Counterfactual with no rainfall anomalies

Source: Authors, using CSO and IMD data.


than the historical trend, improving upon the                                              on Climate Change suggest that precipitation
sustained expansion observed in the 1980s                                                  may likely increase for South Asia but also that
and early 1990s. Thus, contrary to general                                                 spatial and inter-annual variability will intensify
perceptions, there is little evidence to suggest                                           (IPCC 2013).46 Over the short and medium term,
that the policy reforms in the 1990s had a                                                 the more immediate need will be to tackle the
significant adverse impact on agriculture, as                                              shifting trend in weather anomalies, with the
may be inferred from the observed growth                                                   2000s being the worst period for anomalies in
trend. More optimistically, it is tempting to                                              the past century (see Figure 15). In this context,
infer that had rainfall cooperated, the reforms                                            the 1997–2005 growth slowdown heavily
could have ushered in the intended structural                                              underscores the urgency of addressing critical
change, as is often ascribed to the broader                                                policy issues in anticipation of the highly
reforms in the aggregate economy.                                                          uncertain outcomes of climate change:
                                                                                                  [[    The need to improve resilience as the
Implications of Findings                                                                                likelihood of cyclical and random rainfall
                                                                                                        shocks increases, by:
The implications of these findings are more
                                                                                                         Improving the management and
significant than the hypothesized impacts
                                                                                                           efficient use of water resources,
of economic reforms. They dramatically
illustrate the importance of mitigating the
impacts of sustained weather shocks and the                                                46	 The high degree of variability in the predicted patterns of
                                                                                               the El Niño Southern Oscillation, the dominant mode of the
changing climate. The very long term (100-year)                                                predicted precipitation variability, provides low confidence in
predictions by the Intergovernmental Panel                                                     the predictions of outcomes.




     26
                                Chapter 3: Temporal Trends in Agricultural Output and Productivity




        while promoting investments in                   sharing across states and districts in
        sustainable irrigation to weather-               response to clear market signals.
        proof agriculture.
                                                    [[   The need to diversify and stabilize
      Promoting public and private                     sources of income (outside the crop
        investment, as well as greater                   sector), both on-farm (through livestock)
        openness, in introducing suitable                and off-farm (through productive
        technology to mitigate and                       nonfarm employment).
        potentially adapt to climate
        variability.                                [[   The need to improve the effectiveness of
                                                         safety net programs such as the Public
[[   The importance of improved markets                  Distribution System.
     and marketing to allow real-time risk




                                                                                                  27
                                  Spatial Heterogeneity:
4                                 Performance at the Subnational
                                  Level




N
         ational averages provide an aggregate                                       of the variation in productivity is explained by
         picture of progress, but Figure 18 shows                                    district-specific factors (which subsume state
         why a deeper analysis at the subnational                                    factors). Clearly, substantial local effects would
level is essential to understand the variation in                                    be missed by focusing only on the national or
productivity and identify constraints to growth.                                     even the state level.
The time trend (measure of long-term average
growth) alone provides a limited explanation                                         District-level analysis has other advantages. It
for variations in productivity for cereal crops.                                     helps to reveal how broad-based growth is, as
Rainfall and agro-ecological zone indicators                                         well as the drivers of differential performance
explain more of the variation, but their                                             across geographical areas—drivers such
explanatory power is still limited (34 percent).                                     as access to technology, exogenous agro-
State-specific factors make a substantial                                            ecological conditions, the lack of economic
difference, highlighting the key role of state                                       or physical infrastructure, human capital,
policy and institutions. Another substantial part                                    governance, and other variables. After a brief
                                                                                     review of agricultural growth at the state
                                                                                     level, the main analysis in this chapter will
            actors explaining variation in
Figure 18: F
           productivity of cereal crops                                              concentrate on the performance of agriculture
                                                                                     at the district level.
100
 90
 80
 70                                                                                  Agricultural Growth across States
 60
 50                                                                                  Over the long run, agriculture has performed
 40                                                                                  very differently across states (Table 2). Long-
 30                                                                                  term aggregate trends, however, mask
 20
 10
                                                                                     important structural shifts at the state level.
  0                                                                                  Structural transformation appears to have
                                                                                     started in some states, and convergence appears
       a. Trend



                  b. a+Rainfall



                                  c. b+AEZ
                                   speciﬁc


                                             d. b+State
                                                speciﬁc


                                                          e. d+AEZ
                                                           speciﬁc


                                                                     f. c+District
                                                                          speciﬁc




                                                                                     to have started in some state economies
                                                                                     (Binswanger-Mkhize and D’Souza 2011).

Source: Authors, using ICRISAT-NCAP district database.                               Breaking down agricultural performance
Note: AEZ = agro-ecological zone.                                                    by time period throws light on whether



                                                                                                                                   29
Republic of India: Accelerating Agricultural Productivity Growth




Table 2:  Trend growth in Net State Domestic Product, 2004/05–2012/13 (at 2004/2005 prices)
 Low (<2.0%)                                                                          Medium (2-4%)                                                                                            High (>4%)
 State                                               TGR                             State                                                          TGR                                     State                                      TGR
 Keralaa                                            -1.2                   Uttar Pradesh                                                            2.2                 Gujarat         a
                                                                                                                                                                                                                                       4.0
 Himachal Pradesh                                   0.4                    Uttarakhand                                                              2.4                 Maharashtra                                                       4.3
 Punjab                                             1.3                    Odisha                                                                   3.2                 Andhra Pradesh                                                    4.7
 West Bengal                                        1.9                    Tamil Nadu                                                               3.4                 Karnataka                                                         5.1
                                                                           Assam                                                                    3.8                 Jharkhand                                                         5.3
                                                                           Bihar                                                                    3.8                 Rajasthana                                                        5.5
                                                                           Haryana                                                                  3.9                 Chhattisgarh                                                      6.0
                                                                                                                                                                        Madhya Pradesh                                                    6.1
Source:  Authors, using National Accounts Statistics, CSO, as on August 1, 2013.
Note:  TGR = trend growth rate. a Based on data from 2004/05 to 2011/12.

Figure 19: State agriculture growth rates by time period, 1980/81–2011/12
0.12

 0.1

0.08

0.06

0.04

0.02

   0
        Gujarat


                  Orissa


                           Bihar


                                   Madhya Pradesh


                                                     Himachal Pradesh


                                                                        Assam


                                                                                Andhra Pradesh


                                                                                                 Uttar Pradesh


                                                                                                                 Kerala


                                                                                                                          Karnataka


                                                                                                                                      Maharashtra


                                                                                                                                                     Tamil Nadu


                                                                                                                                                                  Haryana


                                                                                                                                                                            Rajahthan


                                                                                                                                                                                            Punjab


                                                                                                                                                                                                     West Bengal


                                                                                                                                                                                                                   Chhattisgarh


                                                                                                                                                                                                                                  Jharkhand


                                                                                                                                                                                                                                                Uttarakhand
-0.02

-0.04

-0.06
                                                                                1980/81-92/93                                     1992/93-99/00                                             1999/00-04/05
                                                                                2004/05-11/12                                     Linear (1980/81-92/93)                                    Linear (2004/05-11/12
Source: Authors, using CSO data on state GDP.


growth rates across states are converging.47                                                                              appear in Figure 19. Focusing attention on the
Growth rates for the major agricultural states                                                                            first and last period, which better represent
                                                                                                                          pre- and post-reform performance and avoid the
47	 Central Statistical Organization (CSO) estimates of State
    Agriculture GDP are used to calculate trend growth rates                                                              confounding effects of rainfall in the slowdown
    in four time periods. The GDP series are available in real                                                            period, it is clear that many of the previously
    prices corresponding to four base periods: 1981/82, 1992/93,
    1999/2000, and 2004/05. The growth rates are estimated                                                                laggard states are participating more fully in
    for each period using the time series denominated to the
    relevant base year, assuming that the weights used to derive
                                                                                                                          the growth process. The trend lines by period
    the series within each period are most appropriate.                                                                   show that wide differences in performance



   30
                                            Chapter 4: Spatial Heterogeneity: Performance at the Subnational Level




across states in the 1980s and early 1990s have                               Real value of agricultural output per
                                                                   Figure 20: 
disappeared after 2004/05, demonstrating a shift                              hectare, 2005–07
toward convergence. In other words, growth has
become more inclusive across states.                                   Chhattisgarh
                                                                           Rajasthan
                                                                              Odisha
                                                                    Madhya Pradesh
Individually, many of the LIS have performed                                    Bihar
better since 2004/05, with growth rates                                   Karnataka
                                                                        Maharashtra
significantly above the base trend line in                                    Assam
                                                                               Kerala
Assam, Bihar, Madhya Pradesh, Odisha, and                                 Jharkhand
Rajasthan. Of the relatively new states created                        Uttar Pradesh
                                                                              Gujrat
from previous LIS in 1990, Jharkhand and                                Uttarakhand
                                                                   Himachal Pradesh
Chhattisgarh have experienced impressive                             Andhra Pradesh
growth in the last period. On the other hand,                           West Bengal
                                                                         Tamil Nadu
growth has gradually slowed in previously good                              Haryana
                                                                              Punjab
performers (Punjab and West Bengal) and in
Kerala. Gujarat and Himachal Pradesh recorded                                            0     20,000 40,000 60,000 80,000 100000
phenomenal performance between 1999 and
                                                                        Vale of Total Agricultural Ouput per Hectare (Rs.), TE 2007
2005, which has subsequently slowed.                                                    (Vertical Line: National Average)

The drivers of growth have differed for                            Source: Kshirsagar and Gautam 2013.
each state, but some commonalities have
emerged. The highlights are the importance of                                 Productivity changes by district,
                                                                   Figure 21: 
diversification into high-value agriculture and                               1990–2007
specialty crops. Examples include Bt cotton in
Gujarat and Maharashtra, soybeans in Madhya
Pradesh, maize and horticulture in Bihar, and
horticulture in some southern states.


Productivity Changes at the
District Level
Figure 20 depicts the variation in productivity
across districts, based on Kumar and Jain’s                        Source: Estimates of district-level productivity from
(2012) estimates.48 Despite some clustering,                       Kumar and Jain 2012.
                                                                   Note: TE = triennium ending.
there is significant variation within states
and agro-ecological zones. Low productivity is
common in arid parts of West India and the                         Figure 21 shows the range of productivity
relatively remote interior districts.                              and changes in productivity between 1990–92
                                                                   and 2006–08. The districts are arranged in
                                                                   increasing level of productivity in the base
48	 Kumar and Jain (2012) estimate crop productivity (value
    of production per hectare) for 388 districts using triennial
                                                                   year. The level of productivity varies widely
    averages ending in 1991–92, 2001–02, and 2007–08.              across districts, being 50 times higher in the



                                                                                                                               31
Republic of India: Accelerating Agricultural Productivity Growth




Figure 22: Changes in inter-district equality (Gini Coefficient), 1991–2007
                                           TE 1991                                 TE 2007                          All-India TE 1991                                 All-India TE 2007
    0.35
     0.3
    0.25
     0.2
    0.15
     0.1
    0.05
        0
            Chhattisgarh

                           Bihar

                                   Assam

                                            UP

                                                 MP

                                                            Odisha

                                                                     Uttarakhand

                                                                                   Jharkhand

                                                                                               Rajasthan




                                                                                                           Punjab

                                                                                                                    HP

                                                                                                                           Gujarat

                                                                                                                                     Kerala

                                                                                                                                              West Bengal

                                                                                                                                                            Haryana

                                                                                                                                                                      AP

                                                                                                                                                                           Tamil Nadu

                                                                                                                                                                                        Maharashtra

                                                                                                                                                                                                      Karnataka
                                                      LIS                                                                                        Non-LIS

Source: Authors, using estimates from Kumar and Jain 2012.


most productive district compared to the least                                                             show less variability but had more modest
productive district.49 Changes in productivity                                                             growth rates.
over time at the district level also vary, but
the trend lines show fairly consistent, broad                                                              Even within states there is considerable inter-
patterns over time.                                                                                        district variability (see Annex 4, Table A.4.1).
                                                                                                           The coefficient of variation across all districts
Districts that were more productive in 1990–91                                                             in India was 58.3 percent in TE 1991,
experienced higher growth throughout the                                                                   increasing slightly to 60 percent in TE 2001
1990s. After 2000, however, districts at the                                                               before declining to 56.9 in TE 2007. The Gini
lower end of the productivity spectrum grew                                                                coefficient shows the same trend, with minor
faster. Despite those advances, large differences                                                          changes in equality over time (Figure 22).
persist. The relative rankings are largely                                                                 One distinct pattern that emerges is that the
preserved over time.                                                                                       LIS have typically seen an increase in cross-
                                                                                                           district inequality (with the notable exception
Variability has increased at the two extremes                                                              of Odisha), whereas the other states, also
of the productivity scale. At the lower end,                                                               previously more productive and diverse, have
some districts experienced greater growth                                                                  seen a decline.50 In other words, as growth
than their peers, whereas some of the                                                                      increases in selected districts within the LIS,
more productive districts performed poorly                                                                 inequality increases as well. In the more
compared to the others, recording lower                                                                    advanced agricultural states, the previously
productivity in 2007 relative to 1990. A vast                                                              lagging districts appear to be catching up,
majority of the districts in the middle range                                                              indicating convergence.
                                                                                                           50	 Punjab is an exception among non-LIS with low inequality
49	 Productivity ranged from Rs. 2,068 per hectare in Barmer                                                   across districts, indicating more uniformly high productivity
    District of Rajasthan to Rs. 107,376 in Karnal District of                                                 levels across districts starting in 1991, which appears to have
    Haryana in the Triennium Ending (TE) 2007–08.                                                              been maintained through time.




   32
                                           Chapter 4: Spatial Heterogeneity: Performance at the Subnational Level




Dynamics of Agricultural                                             rainfall than their faster-growing counterparts,
Productivity at the District Level                                   suggesting that policy rather than exogenous
                                                                     factors may be more important in addressing
The unusual slowdown in Indian agriculture                           inequalities in agricultural productivity
from the late 1990s to mid-2000s, with its                           between districts.
close relation to unusual rainfall patterns,
clouds the analysis of agricultural performance                      The second key issue addressed by the analysis
and its likely drivers between 1990 and 2008.                        is which policy drivers are consistent with
To obtain more robust results, Kshirsagar                            improvements in agricultural productivity at
and Gautam (2013) extend the period of                               the district level. Even in districts previously
the district-level analysis to start from 1970                       characterized by low cereal yields, the main
onward. They address two key issues.51 First,                        driver of productivity has been irrigation
has agricultural growth been geographically                          and the associated adoption of seed-fertilizer
inclusive? The evidence suggests that the                            technology.52 The analysis supporting these
modest sectoral growth over the period was                           findings is presented in the sections that
not shared across districts. Instead agricultural                    follow.
productivity differences between districts
increased between the 1970s and 2000s.                               It is important to note that while the focus
Despite the spread of technology (the green                          of this analysis is agricultural productivity,
revolution technology), data suggest only mild                       inevitably there will be some districts (as well as
convergence. A few districts are “catching                           areas within districts) with few or no prospects
up,” while a substantial number are falling                          for irrigation. Given their agro-ecological
further behind despite positive but low                              endowments (their rainfall, soils, and so on),
levels of growth. Addressing this imbalance                          the scope for productive agriculture may be
remains central to achieve broad-based growth.                       limited. Development strategies thus need to
Importantly, the evidence shows that “laggard”                       be tailored to specific locations, and in some
districts are not at a disadvantage because of                       areas agriculture (or certain types of agriculture,
rainfall levels. On average they received more                       such as cereal production) may not be viable.
                                                                     Livelihood options outside agriculture will be
51	 The analysis uses the district-level database compiled by        essential for sustainable poverty reduction in
    ICRISAT and NCAP. Kumar and Jain (2012) use a subset of that     such areas.
    database in their analysis. The data are from 1970 to 2008 for
    19 major agricultural states. The districts and states created
    through splits and divisions are mapped to the original
    districts as they existed in the 1960s (the “apportioned
    data”). These original administrative districts (and states)     52	 Some caveats should be noted. First, the analysis is based on
    are used in the analysis. Data on output and area are most           econometric estimations using both district and year fixed
    reliable for cereals and a few other crops. For reliability          effects. Nevertheless, regardless of the technique employed,
    the analysis focusses on cereals, but consistency checks             it is not possible to control for all unobservable time-varying
    are made with productivity across all crops. The broader             factors. Second, all measures of agricultural productivity are
    measure is an approximation based on available prices. Only          volatile (partly due to weather, but also likely measurement
    districts with data for the complete period of analysis are          error). For these reasons, decade averages are used for
    retained, totalling 297 of the 317 districts in the database.        some of the analysis, but econometric analysis uses yearly
    These districts accounted for 78 percent of aggregate cereal         data. The analysis is based on district aggregates and not
    production in 2005–07. Kshirsagar and Gautam (2012) show             individual household-level responses. As such, inferences
    that the data provide a very good approximation of long-term         about household behavioral responses will be subject to the
    trends as well as inter-annual fluctuations in the aggregate         ecological fallacy (Freedman 2001; Freedman, Pisani, and
    performance of agriculture in India.                                 Purves 2007).




                                                                                                                                  33
Republic of India: Accelerating Agricultural Productivity Growth




District performance by growth                                           Table 3:  District growth typologies
typology                                                                                         Low growth              High growth
                                                                                                   (<100%)                 (>100%)
The distribution of yields has become more
dispersed over time.53 The progressive                                    Low yields         Laggard                  Low yield/growth
“flattening” of the distribution curve (yields                            (<1.25 MT)
across districts) seen in Figure 23 means that                            High yields        High yield/stable        High yield/
over time more districts have experienced                                 (>1.25 MT)                                  growth
productivity growth, with a larger shift in the                          Source:  Kshirsagar and Gautam 2013.
1970s and 1980s than in the 1990s.                                       Note: “Growth” is defined as percent change in average
                                                                         yields from the 1970s to 2000s. The “high-yield” cutoff is
Although more districts have experienced                                 the 75th percentile in the 1970s.

productivity growth, not all districts have
participated. A significant number seem to be                            performance (Table 3).54 This helps distinguish
stuck at low productivity levels. To identify                            between districts that show slow growth but
the drivers of productivity growth, districts                            do not necessarily have low productivity—for
are categorized by long-run growth typologies,                           example, districts that were more advanced to
based on initial conditions and subsequent                               begin with and likely closer to the prevailing
                                                                         technological frontier, with perhaps fewer
            istribution of yields across districts by
Figure 23: D                                                             prospects of increasing growth in productivity.55
           decade
          1.5                                                            Figure 24 shows the magnitudes of changes
                                                                         in productivity by decade and typology. As
          1
Density




                                                                         noted, most of the analysis here is based on
          .5                                                             cereal productivity; it mirrors the performance
                                                                         based on the broader agricultural productivity
          0                                                              measure (including other crops) providing
                0    1          2           3           4            5
                                                                         confidence that the findings are robust (see
                      Decade Average Cereal Yield (MT/Ha)
                                                                         Annex 4, Figure A.4.2, for the broader measure
                    1970-79   1980-89    1990-99    2000-2007            of productivity).
Source: Kshirsagar and Gautam 2013.


                                                                         54	 Using a short-run growth typology provides qualitatively
53	 For three reasons, levels and changes in cereal yields are               similar findings.
    used for most of the analysis to characterize district-level         55	 An alternative typology considered for this study sought to
    agricultural performance. First, at the district level, historical       exploit the more intuitive rainfed-irrigated categorization
    output data on fruits, vegetables, and spices are not available,         of agriculture, using the proportion of area in a district
    making it difficult to estimate yields or value changes over             that is irrigated. The analysis using that kind of typology is
    time. Results for states that rely heavily on these crops (such          complicated by the fact that irrigation development is not
    as Kerala) will thus be biased downwards. Second, prices for             static; irrigated area can change rapidly over time in many
    many (noncereal) products are unavailable at the district                districts (see Annex 4, Table A.4.2). Another consideration
    (or state) level for many years. Using prices from only a few            is that additional irrigation development may be less of a
    years will likely introduce other biases in the analysis. Third,         priority in areas endowed with good rainfall. Some of the
    the data for cereal yields closely match aggregate cereal                highly irrigated districts are located in the “arid” zone, while
    data used by the GOI, FAO, and the World Development                     a number of districts with little or no irrigation are in the
    Indicators. Nevertheless, consistency checks are made with               “humid” zone, with nontrivial productivity trends, as shown
    the broader productivity measure, as discussed later.                    in Annex 4, Figure A.4.1.




          34
                                                                    Chapter 4: Spatial Heterogeneity: Performance at the Subnational Level




Figure 24: Long-run productivity typology of districts based on cereal yields

                       4
Cereal Yield (MT/ha)




                       3

                       2

                       1

                       0
                           70s        80s    90s    00s         70s      80s        90s        00s         70s        80s    90s        00s         70s     80s    90s     00s
                            High Yield - Growth                      High Yield - Stable                       Low Yield - Growth                         Low Yield - Stable

Source: Kshirsagar and Gautam 2013.

            ispersion of districts by agricultural productivity typology
Figure 25: D
                            100%
                             90%
                             80%
                             70%
                             60%
                             50%
                             40%
                             30%
                             20%
                             10%
                                 0%
                                        OR     AS    MH      BH       MP       GJ         RJ     KT      TN      HP         WB     UP         KE   AP     PB      HR

                                               Low Yield - Stable      High Yield - Stable            Low Yield - Growth         High Yield - Growth

Source: Kshirsagar and Gautam 2013.


Figure 25 maps the districts by growth                                                                Haryana, Uttar Pradesh, and West Bengal),
typology. Except for some districts in Central                                                        accounting for 62 percent of the low-yield/
India, Gujarat, and West Bengal, most of the                                                          growth districts. On the other hand, those
productivity growth has been concentrated                                                             states had just 32 percent of all low-yield
in northern and southern districts. Table                                                             districts in the 1970s. In contrast, Assam,
4 more clearly shows the distribution of                                                              Bihar, Maharashtra, Madhya Pradesh, and
districts by typology and state. Of the districts                                                     Odisha had a disproportionately large fraction
with initially low yields (225 of the total                                                           of districts that failed to raise productivity
297 districts), less than half had doubled                                                            appreciably over three decades despite very
productivity by the 2000s. Of these, 86                                                               low productivity in the 1970s.
percent were from just five states. Within
states, in only four were the vast majority                                                           The analysis by growth typology provides a
of districts doing well (Andhra Pradesh,                                                              number of key insights into the patterns and



                                                                                                                                                                           35
Republic of India: Accelerating Agricultural Productivity Growth




Table 4:  District growth typologies by state (long run, 1970–2007)
                                              High yield/        High yield/            Low yield/       Low yield/
    State                                      growth              stable                growth            stable               Total
    Andhra Pradesh                                  5                  2                    12                 1                  20
    Assam                                           0                  0                      0               10                  10
    Bihar                                           0                  3                      0                8                  11
    Gujarat                                         0                  0                      7               11                  18
    Haryana                                         2                  1                      4                0                   7
    Himachal Pradesh                                0                  7                      0                2                   9
    Karnataka                                       0                 11                      1                7                  19
    Kerala                                          0                  9                      0                1                  10
    Madhya Pradesh                                  0                  0                    16                27                  43
    Maharashtra                                     0                  4                      2               19                  25
    Odisha                                          0                  0                      0               13                  13
    Punjab                                          1                 10                      0                0                  11
    Rajasthan                                       0                  0                    14                12                  26
    Tamil Nadu                                      1                  7                      0                4                  12
    Uttar Pradesh                                   3                  3                    37                 5                  48
    West Bengal                                     1                  2                    10                 2                  15
    Total                                          13                 59                   103               122                 297
Source:  Kshirsagar and Gautam 2013.


Figure 26: Share of cereal area planted to high-yielding varieties
            1

            .8
HYV Share




            .6

            .4

            .2

            0
                 70s   80s    90s       00s     70s   80s     90s       00s       70s     80s   90s   00s       70s      80s    90s     00s
                  High Yield - Growth             High Yield - Stable               Low Yield - Growth                Low Yield - Stable

Source: Kshirsagar and Gautam 2013.


sources of productivity growth across districts.                              have now diffused widely. Use of HYVs grew
First, improved seeds alone do not prove to be a                              significantly in the laggard districts,
sufficient driver of growth. High-yielding varieties                          yet no corresponding productivity
(HYVs) were more commonly planted in the                                      improvement occurred (Figure 26). Given
high-yield districts in the 1970s at the start                                the rapid rate of adoption across all types
of the green revolution, but improved seeds                                   of districts, HYVs do not offer a sufficient



            36
                                                                                Chapter 4: Spatial Heterogeneity: Performance at the Subnational Level




 Figure 27: Irrigation intensity across district typologies by decade

                               .8
Irrigation Intensity




                               .6

                               .4

                               .2

                               0
                                        70s         80s         90s     00s     70s     80s     90s         00s         70s      80s     90s    00s    70s     80s     90s    00s
                                         High Yield - Growth                      High Yield - Stable                     Low Yield - Growth             Low Yield - Stable

 Source: Kshirsagar and Gautam 2013.


 Figure 28: Fertilizer intensity across district typologies by decade
Fertllizer Intensity (Kg/ha)




                                300


                                200


                                100


                                    0
                                              70s         80s     90s     00s     70s     80s         90s     00s          70s     80s    90s    00s    70s     80s    90s    00s
                                               High Yield - Growth                    High Yield - Stable                     Low Yield - Growth              Low Yield - Stable

 Source: Kshirsagar and Gautam 2013.


 explanation for the differences in productivity                                                                    a convincing explanation. On average, high- and
 growth.56                                                                                                          low-yield districts where productivity increased
                                                                                                                    most received less rainfall on average, and in
 Second, agro-ecological conditions alone also do not                                                               fact suffered greater negative anomalies in the
 sufficiently account for the disparities in productivity                                                           2000s relative to their more stable counterparts
 among districts. A little over half of the low-yield                                                               (Annex 4, Figures A.4.3 and A.4.4).
 districts in the Semi-Arid Tropical Zone did not
 experience significant growth—but one-third                                                                        Irrigation and fertilizer set the growth districts apart
 did. Similarly, a majority of low-yield districts in                                                               from the others. High-yield districts had better
 the Temperate Zone saw yields grow, yet almost                                                                     irrigation to begin with (Figure 27), but a key
 one-quarter did not (Annex 4, Table A.4.3). Third,                                                                 difference between the growth and stable
 rainfall patterns and anomalies also do not provide                                                                districts is the growth in irrigation intensity,
 56	 This finding needs to be examined at a lower level of
                                                                                                                    especially in the low-yield/growth districts.
     aggregation (using household or plot-level data). There may                                                    Fertilizer use has also grown much faster in
     be substantial differences in the efficacy of HYV seed, or the
     varieties that are available may not be suited to the agro-
                                                                                                                    the growth districts, likely influenced by
     ecological conditions in particular districts.                                                                 irrigation (Figure 28).



                                                                                                                                                                              37
Republic of India: Accelerating Agricultural Productivity Growth




An analysis of growth typology by the                                   through the use of tanks. The high-yield/
dynamics of irrigation confirms the strong                              growth districts initially had a large area
association of irrigation in the growth                                 under canal irrigation (in absolute terms and
districts and provides additional insights                              relative to other districts), but their share in
(see Annex 4, Table A.4.4). The majority of                             area irrigated has declined consistently in each
the growth districts are concentrated in                                subsequent decade.
high-irrigation districts or those that have
developed irrigation over time.57 However,                              One of two key questions posed earlier was the degree
there are important exceptions. About one-                              to which growth has been inclusive. The fact that
third of the high-yield/stable districts had low                        more districts are participating in growth
irrigation and still have low irrigation. Almost                        may suggest convergence across districts. But
all of these are in the “humid” zone, either                            despite the growth in the laggard districts,
in the coastal areas or the ghats of Kerala,                            the difference between the laggards and
Karnataka, and Maharashtra. At the same time,                           high performers is rising (Figure 29). Clearly,
more than half of the laggard (low-yield/stable)                        convergence in cereal productivity across
districts either had a high level of irrigation                         districts is modest and limited to a few districts.
or developed irrigation capacity over time                              When the analysis included all crops for which
but have not seen significant increases in                              data are available, not just cereals, the results
cereal or overall productivity.58 Importantly,                          suggest divergence rather than convergence.
a substantial number of districts (50 of 295,                           Except for a handful of districts where
or about 30 percent) are in the laggard and                             productivity was initially very low, there is little
low-irrigation category. They have developed                            evidence to support convergence (Annex 4,
no significant irrigation capacity so far, and                          Figure A.4.6). The more advanced districts
the micro-level data on their potential for                             appear to be distancing themselves from the
irrigation development are not available.                               others.

Changes in the mode of irrigation are also                              The persistence of a large number of
revealing (Annex 4, Figure A.4.5). The use                              laggard districts, confirmed by the lack of
of pumps and wells to extract groundwater                               convergence across districts, suggests a need
increased consistently over the decades, with                           for more detailed assessments of their agro-
extraordinary increases in the low-yield/growth                         ecological potential. In some districts, or
districts. Irrigation intensity in laggard districts                    even in parts of districts, it is likely that the
and high-yield/stable districts increased mostly                        natural environment may not be conducive
                                                                        for rapid productivity growth. Development
57	 Districts are categorized in terms of their irrigation dynamics
    for this analysis. The “low” group is the set of districts that     strategies thus need to be tailored to specific
    had a low level of irrigation (less than 25 percent cultivated      locations, based on detailed analyses of the
    area) in the 1970s and continues to have low irrigation. The
    moderate/dynamic group comprises districts that had low or          types of agricultural activities that may be
    moderate (25–50 percent area) levels of irrigation in 1970s         viable. It is inevitable that in some locations,
    but have since increased irrigated area. The “high” group
    has had a high level of irrigation (more than 50 percent area)
                                                                        sustainable income growth and poverty
    since the 1970s.                                                    reduction objectives may be better served by
58	 Some level of crop diversification (toward high-value crops
    such as horticultural crops) is evident in all districts, but has
                                                                        supporting avenues toward livelihoods outside
    been the highest in low-yield/growth districts.                     of agriculture.



   38
                                                            Chapter 4: Spatial Heterogeneity: Performance at the Subnational Level




            hanges in district agricultural productivity
Figure 29: C
                               5



                               4
   Avg. Cereal Yield : 2000s




                               3



                               2



                               1



                               0

                                   0                               1                                     2                                   3
                                                                           Avg. Cereal Yield : 1970s

                                       Fitted Values   High-Yield Growth        High-Yield Stable      Low-Yield Growth   Low-Yield Stable

Source: Kshirsagar and Gautam 2013.


Drivers of productivity growth                                                         of new technology and nontraditional crops in
                                                                                       driving productivity and growth.
Productivity can grow through intensification
or through switching to crops that are more
                                                                                       Data for indicators that explain productivity
productive (in value per hectare). Area changes
                                                                                       changes are limited. Despite poor quality and
show some shifts, but no obvious patterns
                                                                                       missing data for different years and districts,
emerge across district types (Annex 4, Table
                                                                                       the main changes observed are presented
A.4.6). The area under cereals has declined only
marginally. Most cropped area remains under                                            in Table 5.
cereals in all types of districts, and cereal area is
still highest in the high-yield/stable group, the                                      Markets per capita have declined in low-yield
original green revolution strongholds. While                                           districts relative to the other districts. This
the area under horticulture grew in all districts,                                     change may represent a reduction in incentives
it remains relatively lower in the low-yield                                           for producers, constraining productivity
districts, and the share of pulses and oilseeds                                        growth. Road density was low in the early
remains relatively higher. The low-yield/growth                                        years in low-yield districts, but has since
districts have departed from the laggards in                                           grown faster than in the other districts. But
that they have reduced their area under pulses                                         the increased investment in infrastructure
and switched to higher-value crops, including                                          has not yet translated to productivity growth
cotton, horticultural crops, and newer oilseeds                                        in the laggard (low-yield and stable) districts.
(soybeans)—clearly highlighting the importance                                         Low-yield districts had relatively lower levels



                                                                                                                                             39
Republic of India: Accelerating Agricultural Productivity Growth




Table 5:  Levels and changes in key indicators of changes in agricultural productivity
                    High yield/growth     High yield/stable     Low yield/growth       Low yield/stable
                                    %                     %                    %                      %
Area share        1970s 2000s change 1970s 2000s change 1970s 2000s change 1970s 2000s change
Markets/capita     0.02 0.02       -15.9 0.02 0.02        9.6 0.01   0.01     -20.1 0.02 0.01        -25.2
Road density        3.6    7.3    100.9   5.7 10.1       78.8  2.1     5.2   150.0    2.9    7.1    150.4
Fraction rural     0.79 0.71       -10.1 0.81 0.72      -11.3 0.82   0.76       -6.7 0.86 0.79         -7.6
Rural literacy     0.22 0.52      131.0 0.33 0.62        87.2 0.19   0.47    152.8 0.22 0.49        123.6
Source:  Kshirsagar and Gautam 2013.


of urbanization (higher fraction of rural                       reorienting investments to promote
population) to begin with; they have seen                       growth in noncereal crops in low-
slower change in this indicator, with the                       productivity areas.
laggard districts at the greatest disadvantage.
                                                           [[   Fertilizer shows a consistent impact on
Finally, literacy levels have increased but
                                                                cereal productivity across all districts,
remain marginally low in the low-yield districts.
                                                                but its impact on noncereal crops is
                                                                limited, possibly reflecting their higher
The descriptive analysis above is substantiated
                                                                risk or inappropriate use.
by econometric estimates of the determinants
of productivity growth (Annex 4, Tables A.4.7–             [[   Roads appear to have a greater impact
A.4.10). In brief, increases in irrigation use appear           in high-yield areas (across crops),
to have driven productivity growth in districts                 confirming the importance of roads
that had low yields in the 1970s, whereas growth                for productivity growth. This result
in the more mature districts was driven by a                    suggests that increased road density
more balanced constellation of factors:                         (noted earlier) has not yet translated
                                                                to productivity in low-yield areas,
    [[   Rainfall is a significant determinant (for
                                                                calling attention to complementary
         all crops) in the low-yield districts, as is
                                                                productivity-enhancing investments
         irrigation. The insignificance of rainfall
                                                                which may be lacking.
         in high-yield districts reinforces the
         importance of irrigation in mitigating            [[   To pick up the location-specific time-
         rainfall shocks—confirmed by the                       invariant factors, cross-sectional
         insignificance of rainfall in the high-                regressions show high sensitivity of
         irrigation districts. At the same time,                productivity to both mean rainfall
         rainfall’s significant impact in low-                  (location effects) and the variability in
         irrigation districts indicates quite clearly           rainfall (risk). Analyses at a lower level
         that the potential payoffs to irrigation               of aggregation are needed to uncover
         development in low-yield/low-irrigation                causal mechanisms, but these findings
         districts are significant. Importantly,                suggest that within states and agro-
         irrigation also drove productivity in                  ecological zones, weather risk has a
         noncereal crops in high-yield/growth                   detrimental impact on productivity an
         districts, implying a potential role for               growth.



  40
                                 Chapter 4: Spatial Heterogeneity: Performance at the Subnational Level




   [[   Irrigation intensity and market                       districts increasing between 1970 and
        density have strong positive impacts                  2007. The analysis, however, defies the
        on productivity, after accounting                     search for a “silver bullet.” For example,
        for rainfall and agro-ecological                      investment in roads alone has not yet
        diversity. But their influence becomes                boosted the laggard districts, but it does
        statistically insignificant when state                show significant impact in high-yield
        effects are introduced, suggesting high               districts. Similarly technology alone has
        a correlation with state-level effects,               not spurred growth. Complementary
        highlighting the important role of                    investments in irrigation and market
        state policies governing investments                  intensity show large impacts,
        and institutions as key drivers of                    importantly through diversification,
        productivity growth.                                  distinguishing better performers from
                                                              laggard districts. But the effectiveness
                                                              of these instruments varies by state,
Implications of the Subnational                               highlighting the critical importance
Analysis                                                      of state policies governing institutions
                                                              and investments to promote inclusive
The main messages emerging from the analysis
                                                              growth.
of agricultural performance at the subnational
level are:                                               �	   Water and productivity. Districts that
                                                              have increased productivity most
   �	   Policy reform, risk and agricultural
                                                              rapidly since the 1970s did so with
        performance. Consistent with the
                                                              significant increases in irrigation—
        previous finding that the policy
                                                              mostly groundwater extraction—and
        reforms of the 1990s likely had a
                                                              closely correlated adoption of HYVs
        stimulatory impact on productivity,
                                                              and fertilizer. Groundwater extraction
        several previously laggard states have
                                                              is already unsustainable in many parts
        significantly improved growth after
                                                              of the country,59 rapidly reducing
        2004–05 under relatively normal
                                                              aquifers60 and threatening ecological
        rainfall. The slowing of growth in states
                                                              sustainability.61 A direct link to energy
        that previously were highly productive
        is a cause for concern, however. These
                                                     59	 While the design and implementation of a “rational water
        conclusions are reinforced by the                policy” is a national priority—see, for example, Ahluwalia’s
        empirical importance of the detrimental          (2011) discussion of such a policy in the context of India’s
                                                         Twelfth Five Year Plan—several relevant actions involve state-
        impact of weather risk on productivity           level decisions in a context in which inter-state externalities
        and long-term growth prospects for               are large and a consensus is still emerging.
                                                     60	 Confirmed in recent publications from the Groundwater
        agriculture.                                     Board. See also Rodell, Velicogna, and Famiglietti (2009) on
                                                         effects in Punjab, Haryana, and Rajasthan between 2002
   �	   Promoting broad-based growth. Agricultural       and 2008.
        productivity has grown unevenly across       61	 M.S. Swaminathan has often referred to the use of subsidies
                                                         for excessive groundwater abstraction for irrigation as
        districts—even within states and agro-           ecological suicide or “ecocide.” See, for example, the
        ecological zones. A significant share            interview reported in the Frontline magazine (Volume 17,
                                                         Issue 11, May 27-June 9, 2000) accessible at: http://www.
        of districts did not participate fully           frontline.in/navigation/?type=static&page=flonnet&rdurl=fl17
        in growth, with disparities between              11/17110770.htm.




                                                                                                                 41
Republic of India: Accelerating Agricultural Productivity Growth




       subsidies has also been shown by a                  good access to water (or have developed
       recent rigorous study (as discussed later).         it over time), but they have not
       At the same time, mere availability of              managed to improve their performance,
       water does not guarantee success. More              pointing to other binding constraints to
       than half of the laggard districts have             productivity growth.




  42
5                   Structural Change




T
      he worrisome and growing gap in                       in the more labor-intensive manufacturing
      productivity between agricultural                     sector (Bosworth, Collins, and Virmani 2007;
      and nonagricultural workers—a key                     Bosworth and Collins 2008; Binswanger-
indicator of economic well-being—traces                     Mkhize and D’Souza 2011; World Bank 2012a;
back to the slow transition of labor out of                 GOI 2013a). The stylized process of structural
agriculture. Large numbers of people remain                 transformation, in which labor exits from
“stuck” in agriculture because growth and                   agriculture as an economy develops (Box 1),
job creation have not been sufficiently rapid               appears to be occurring more slowly in India

Box 1: Structural transformation of economies

Based on historical evidence, several distinct processes are associated with structural transformation
of economies: (1) the share of agriculture in GDP declines, (2) the share of agriculture in employment
declines, (3) rural–urban migration increases, (4) the service and manufacturing sectors grow rapidly, and
(5) a demographic transition occurs with a reduction in population growth rates.a The final outcome of
structural transformation is a state in which differences in labor productivity between the agricultural and
nonagricultural sectors disappear.
Growth in agricultural productivity is fundamental for structural transformation to proceed at a steady
pace, because as growth in other sectors accelerates, agriculture’s share in GDP declines rapidly, yet
the share of the population deriving a living from the agricultural sector declines more slowly. Income
inequalities across sectors thus tend to widen initially with a concentration of poverty in agriculture.
Eventually a turning point is reached, and labor productivities in the agricultural and nonagricultural
sectors begin to converge.b
For the turning point to occur, agricultural productivity must increase—partial (land and labor) but
also total factor productivity. As structural transformation proceeds, labor will move out of agriculture,
making it critical that the remaining resources become more productive to provide the food, savings, and
investments for development of the nonagricultural sector. Unless that happens, inter-sectoral terms of
trade can move in favor of agriculture, raising food prices, arresting transformation, and even threatening
sociopolitical stability.c These insights from theory and historical experience are pertinent to India, which
has experienced a strong movement in the terms of trade in favor of agriculture, less-than-desirable poverty
reduction, and a disproportionate share of the labor force in agriculture despite the rapid economic growth
of the past two decades.
         Lele et al.2013, using Kuznets 1955, 1966; Chenery and Taylor 1968, Chenery et al 1974, Chenery and Syrquin
Source: a
        1975; Timmer 2009. b Lewis 1954; Johnston and Mellor 1961; Timmer 2009. c Kuznets 1955, 1966; Lewis 1954;
        Johnston and Mellor 1961; Lele and Mellor 1981; Mellor and Lele 1973.




                                                                                                                 43
Republic of India: Accelerating Agricultural Productivity Growth




than elsewhere (Hazell et al. 2011; Foster and       countries.62 In 1961, condition in the three
Rozensweig 2010). India’s transformation             Asian countries were comparable. Since then,
has also been characterized as “stunted”             China and Indonesia have achieved more
(Binswanger-Mkhize and D’Souza 2011),                success in reducing poverty and hunger. Brazil
with the exiting labor moving primarily              was relatively more advanced from the start,
into the rural nonfarm and informal sector           and it has continued to perform well since.
and becoming increasingly “casualized”               India’s agricultural productivity growth has
(World Bank 2010; Lanjouw and Murgai 2009,           been slower than that of the other countries.
Himanshu et al. 2010).                               How well have each of these countries done
                                                     in agriculture to meet their development
This atypical demographic transition places a        challenges? Going forward, what are the
heavy tax on rural well-being and prosperity.        implications and priorities for policy and
Agricultural labor productivity as a ratio of land   investment in India?
productivity has declined by almost 1 percent
per year between 1980 and 2009. In other             Divergent patterns of structural
words, physical productivity growth (modest          transformation
as it is) has not commensurately improved
the standard of living of those engaged in           Building on earlier studies of structural
agriculture.                                         transformation, Lele et al. (2013) assess the
                                                     agricultural performance of the four countries
This chapter examines India’s structural             against the background of an analysis of
transformation. It compares India’s experience       panel data for 109 developed and developing
to that of other countries to identify issues        countries between 1980 and 2009. The data
for consideration as India moves forward. An         reaffirm the stylized facts about structural
equally useful aperture into structural change       transformation and development. The shares
in India comes from village-level analysis. A        of value added and employment in agriculture
longitudinal study provides insights into the        decline as per capita incomes rise (albeit at
development process in one of the poorest            a decelerating rate). The total value added
states, Bihar.                                       in agriculture and value added per worker
                                                     (a measure of labor productivity) increase
                                                     with per capita income. Significantly, labor
India’s Structural                                   productivity rises faster when the developed
Transformation in an                                 countries are included in the analysis, but
International Perspective                            it rises more slowly when only developing

Lele et al. (2013) compare India’s performance
                                                     62	 Brazil and Indonesia, each with populations of around 200
with that of three other large developing                million or more, are rich in agricultural, forest, and other
countries—Brazil, China, and Indonesia.                  natural resources, and they are already major exporters of
                                                         food and agricultural products. China and India, each with
Together the four countries (henceforth BIIC)            a population exceeding one billion, cannot bring more
contain 44 percent of the global population              land under cultivation, and future agricultural growth
                                                         is contingent on more efficient use of a limited natural
and produce slightly more than one-third of              resource base. The increasing scarcity of land and water are
all cereals. All are rapidly growing emerging            two particularly binding constraints.




  44
                                                                                   Chapter 5: Structural Change




Table 6:  Turning points for BIIC
                       Per capita income, 2010     2004-08 growth rate           2009-10 growth rate
 Country                      (2000 US$)         (Number of years needed)      (Number of years needed)
 Brazil                         4699.4                              Already reached
 China                          2425.5                      4                              5
 India                           822.8                    23                             21
 Indonesia                      1143.8                    27                             29
Source:  Lele et al. 2013.


countries are included (88 total). This                reach their respective turning points (Table 6).63
discrepancy may signal that the development            The turning points are based on the per capita
context is changing; as a result, today’s              income in each country in 2010 and vary
developing countries may find that raising             according to the assumptions about growth
labor productivity is now a more challenging           rate of the per capita income. Based on recent
task than it was for countries that developed in       growth rates (2004–08), Brazil has passed its
the past.                                              turning point. China is almost there, thanks
                                                       to rapid growth in agricultural value-added
The gap between the share of value added               despite having a higher share of its population
originating in agriculture and the share of            in agriculture than India. In contrast, India and
employment in agriculture is an indicator for          Indonesia will take longer to reach their turning
the level and pace of structural transformation.       points (about two decades for India and almost
It reflects the degree of convergence in the per       three for Indonesia).
capita productivity and incomes of workers
in agriculture and the rest of the economy.            Overall, Asia seems to be behaving much
An earlier study (Timmer and Akkus 2008)               as Kuznets predicted, with increasing inter-
found that rapidly growing economies (current          sector duality in initial stages (Figure 30).
developed countries) are reaching the turning          Labor productivity grew much faster in the
point (when labor productivities begin to              nonagricultural sectors than in agriculture in
converge across sectors) at increasingly later         China. In Indonesia the difference increased
stages in their structural transformation. The         up to the time of the Asian financial crisis
implication is that industry and services are          but has since been stable. Nonagricultural
finding it increasingly difficult to absorb labor      labor productivity rose in India as well (but
out of agriculture—a phenomenon that India             less steeply than in China); India’s growth in
too seems to be experiencing.                          agricultural labor productivity was relatively
                                                       slower, and the gap has continued to widen.
Using updated data and more detailed analysis,         On the other hand, it declined continuously in
Lele et al. (2013) find that different regions         Brazil over the 1980–2009 study period, owing
behave differently and in a much more complex
fashion. For the sample of developing countries        63	 Using an alternative specification of the annual fixed effects,
                                                           they find that the turning points would be approximately a
only, they estimate the number of years it will            decade later than reported in Table 7, which is based on a
take Brazil, India, Indonesia, and China to                regression using decade fixed effects.




                                                                                                                    45
Republic of India: Accelerating Agricultural Productivity Growth




            atio of value added per worker
Figure 30: R                                                      Share of value added minus share of
                                                       Figure 31: 
           (BIIC), nonagriculture/agriculture:                    employment in agriculture, residuals
           1980–2009                                              (BIIC): 1980–2009
18                                                                  0.2
16                                                                 0.15
14
                                                                    0.1
12
                                                                   0.05
10




                                                       Residuals
 8                                                                    0                                                     2000
                                                                           1980            1985           1990       1995
 6                                                                 -0.05                                                           2005
 4
                                                                    -0.1
 2
                                                                   -0.15
 0
                                                                    -0.2
     1980
     1981
     1982
     1983
     1984
     1985
     1986
     1987
     1988
     1989
     1990
     1991
     1992
     1993
     1994
     1995
     1996
     1997
     1998
     1999
     2000
     2001
     2002
     2003
     2004
     2005
     2006
     2007
     2008
     2009                                                          -0.25
          Brazil       China   India       Indonesia                              Brazil          China          India      Indonesia

Source: Lele et al. 2013.                              Source: Lele et al. 2013.


to a combination of very rapid growth of labor         the pattern seen in the 88 developing countries
productivity in agriculture (fastest among             in the sample (Figure 31). Progress has slowed
the BIICs) as well as a surprising decline in          since about 2000, however, indicating that
nonagricultural labor productivity.                    labor is not exiting agriculture as fast as
                                                       agriculture’s share of value added is falling.
The share of the labor force in agriculture            China and Brazil are outliers, reflecting their
also shows divergent patterns across the four          trends in labor shares, while Indonesia has
countries. China is an outlier, in that it appears     remained stable and close to the cross-country
to be losing agricultural labor much more              average.
slowly than predicted by the cross-country
experience. Brazil is an outlier in the other          Although China retained a large percentage
direction, shedding labor much more rapidly            of workers in agriculture (concerns about
than predicted. India and Indonesia behave             data on China’s agricultural labor force
like other developing countries. The share of          notwithstanding),64 it is clear that rapid growth
agricultural value added has also been close to        in agriculture has led to substantial gains in
the model predictions for India and Indonesia,         valued added per worker in China. Growth in
whereas agriculture in China and Brazil has            labor productivity is lower in China than in
grown rapidly and retained a significantly larger      Brazil, but it is substantially higher than in
share of the overall economy compared to the           India and Indonesia. Sluggish growth in India’s
other developing countries.                            agricultural productivity, despite a lower share
                                                       of the labor force in agriculture, has meant that
India’s progress on structural transformation—         India’s agricultural population has seen less
measured as the difference between
                                                       64	 There is an ongoing debate about the quality of the labor
agriculture’s shares of value added and the
                                                           data on China. Some have argued that labor in agriculture is
overall labor force—has been consistent with               significantly less than reported.




     46
                                                                                               Chapter 5: Structural Change




Figure 32: Decomposition of output growth, BIIC, over the long and short run
                         Long Run: 1960-2009                                                 Short Run: 2000-2009

5.0%                                                                8.0%

4.0%                                                                6.0%
                                                                            4.0%
           2.2%           2.0%
3.0%                                     1.5%                       4.0%
                                                                                                           3.7%
2.0%                                                     1.2%                                                           2.1%
                                                                    2.0%                      2.8%

1.0%                                                                0.0%
                                                                            Brazil            China       Indonesia         India
0.0%                                                                -2.0%

           Brazil         China        Indonesia         India
-1.0%                                                               -4.0%

              Irrig    New land    Input/Area      TFP                               Irrig    New land   Input/Area   TFP

Source: Authors, using data from Fuglie 2012.


improvement in its standard of living relative                      indicating significantly improved efficiency in
to China.65                                                         the use of inputs. TFP growth has been slowest
                                                                    in India, though with some improvement in
The role of total factor productivity                               the recent decade. And in significant contrast to
                                                                    India, Brazil’s input use has declined, while for
Even if labor is slow to exit from agriculture                      China and Indonesia the contribution of inputs
in India, labor productivity could still be made                    to output growth has been virtually nil.
to rise—as it has in China—by increasing
land productivity or using inputs more                              Productivity growth (net of inputs) can be
efficiently. Figure 32 shows sources of growth                      enhanced through higher yields (better
in agricultural output for each country over                        technology) or diversification into higher-value
the long (1960–2009) and short (2000–09) runs.                      crops. Brazil diversified heavily into soybeans
Land-abundant Brazil and Indonesia expanded                         and livestock, for example, and Indonesia
agricultural area more than India and China.                        into oil palm. China diversified significantly,
China has intensified input use over the                            with a rapid expansion in livestock and high-
long run, whereas India relied on inputs and                        value fruits and vegetables. India has started
irrigation to increase output. In Brazil, China,                    to diversify, but agriculture remains heavily
and Indonesia, total factor productivity (TFP)—a                    concentrated in food grains. Yields of wheat,
comprehensive measure of how effectively                            sugarcane, fruits, and vegetables have grown
inputs are used in production—has been the                          relatively well since the 1980s, but yields of
primary source of growth in agricultural output,                    other crops are significantly lower than in the
                                                                    other three countries.
65	 Agricultural terms of trade in all three Asian countries
    show similar trends since 1980, indicating that shifts in
    terms of trade alone do not explain the significantly better    How do TFP growth rates compare across BIIC
    performance in living standards in China. From 1980 to the
    mid-1990s, Brazil saw a rapid decline from very high terms of
                                                                    and selected regions? India’s performance was
    trade in favor of agriculture, which have since stabilized.     comparable to that of Brazil, Indonesia, and



                                                                                                                              47
Republic of India: Accelerating Agricultural Productivity Growth




China to about 1980 (Figure 33). Yet despite the               China’s industrial reforms in the early 1990s
benefits of the green revolution, the long-run                 sustained the TFP boost from the earlier more
TFP growth has been lower in India than in the                 fundamental reforms of the 1980s, enabling
other countries and regions—except for sub-                    labor to move out of agriculture to help
Saharan Africa (as a whole).                                   raise labor productivity. Continuous capital
                                                               investment and technical change sustained
Nin-Pratt, Yu and Fan (2009) analyze the                       output per worker in addition to promoting
contribution of technical change and efficiency                further growth in agriculture. In India, after the
in explaining TFP growth in India and China                    initial boost of technology in the 1970s reversed
between 1961 to 2006 (Figure 34). Their                        the declining TFP of the 1960s, growth has been
findings provide an insightful comparison of                   sustained, but comparatively slow, with much
each country’s respective pre- and post-reform                 slower movement of workers out of agriculture.
performance (using 1980 as the reform cut-off
for China and 1991 for India). They reiterate the              While the impact of policy reforms in India was
widely recognized importance of a conducive                    likely stifled by rainfall shocks as previously
policy environment for efficiency and growth.                  highlighted, the role of the current agricultural
                                                               policy environment nevertheless remains
Both countries benefited from positive technical               important. How important this may be is
change through the green revolution, irrespective              demonstrated by an independent comparative
of economic reforms, underscoring the critical role            assessment of the policy environment provided
of technology in driving productivity. But with                by the World Economic Forum’s Global
more fundamental institutional reforms, the                    Competitiveness Index (GCI). One component of
movement of workers into rural enterprises                     the GCI is an index on Agricultural Policy Costs
(small manufacturing), fiscal decentralization,                (APC), which is a window on the views of global
and the steady introduction of market                          and national economic and business leaders. Of
incentives through consistent and sustained                    142 countries surveyed for the APC, India ranks
reforms, Chinese agriculture has become much                   79th, just below the median for all countries
more efficient.                                                and significantly below Brazil, Indonesia, and

            FP growth in BIIC and selected
Figure 33: T                                                              Drivers of TFP, India and China,
                                                               Figure 34: 
           regions                                                        1961–2006
330                                                            4                        3.4
            SSA
            AM ERICA, North                                    3
280                                                                                   2.5
            Indonesia                                          2
            MENA                                                                                                       0.3
                                                               1                                      0.2
230                                                                    1.1            0.9
            China                                              0
            Brazil                                                  Pre-reform     Post-reform                   Post-reform
                                                               -1                                 Pre-reform         0.2
180         India                                                       -3.8
                                                               -2              China                  -0.6   India

130                                                            -3
                                                               -4
                                                                       -2.7
 80                                                            -5
      1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 2009           Technical Change          Technical Efﬁciency         TFP
Source: Authors, using data from Fuglie 2012.                  Source: Based on data from Nin-Pratt, Yu and Fan 2009.




  48
                                                                                                     Chapter 5: Structural Change




           The World Economic Forum’s Global
Figure 35:                                                                and tested strategy for promoting growth.
           Competitiveness Index                                           Brazil, India, and China all invested heavily
 6                                                                         in technology and research. They are among
                  Agricultural Policy Cost Index, 1 to 7 (best)
5.5                                                                        the largest investors in public research, from
 5                                                                         which they have benefited significantly. China
4.5
       India’s rank: 79 out of 142                                         has made sustained and substantially more
                                                                           investments, and its investments have also
 4
                                                                           been more effective. Despite operating in a
3.5
                                                                           different environment, Brazil has also invested
 3                                                                         heavily and very effectively in technology.
2.5                                                                        Through Embrapa66 (its national research
 2                                                                         organization) and farmer-led innovation, Brazil
                                                                           has emerged as a global leader in agricultural
                             India




                                                          Brazil

                                                                   China
                                              Indonesia




                                                                           exports (Rada and Valdes 2012). Technology
                                                                           has been pivotal in Indonesia as well. Even
Source: World Economic Forum 2012.                                         though its domestic research has not been
                                                                           as productive as that of the other countries,
China in terms of the policy environment for                               Indonesia benefited from a policy of open trade,
agriculture (Figure 35). The big difference in                             enabling imported technology from Malaysia
rankings helps explain the significant difference                          to unleash extraordinary development in palm
in efficiency as a driver of productivity,                                 oil production and processing (Rada and Fuglie
discussed earlier.                                                         2012).

                                                                           Diversification. A second major driver of
Drivers of growth in the
                                                                           growth in productivity was the diversification
comparator countries: Implications                                         of agriculture, both for domestic and export
for India                                                                  markets, supported by appropriate technology,
The comparator countries differ significantly                              policy, and institutional capital. Brazil,
in their physical endowments and political and                             Indonesia, and China all benefited significantly
institutional systems, but they offer valuable                             more than India from openness in trade. India,
perspectives on Indian agriculture. The “small-                            driven primarily by food security, has only
scale and efficient” experiences of China and                              recently and slowly started to diversify out
Indonesia are more suitable to India than the                              of cereals. China and Brazil diversified out of
“large-scale, mechanized, and efficient” model                             cereals and into livestock; Brazil and Indonesia
of Brazil, but important insights emerge for                               moved heavily into the production and export
both environments.                                                         of soybeans and palm oil—all major sources of
                                                                           growth.
Investment in technology. The first—and
perhaps most important—factor driving                                      Enabling policy environment. China
productivity in all four countries, irrespective of                        and Brazil have a more predictable overall
the policy environment, has been technology.
Investing in technology is a globally proven                               66	 Empresa Brasileira de Pesquisa Agropecuária.




                                                                                                                              49
Republic of India: Accelerating Agricultural Productivity Growth




enabling policy environment for agriculture,        2009–11 with three rounds of panel surveys in
with a strong record of implementation              36 villages of Bihar—among the poorest and
and continuous innovation in public sector          least-developed states in India. Socioeconomic
management for agriculture and rural                patterns shifted dramatically over that period,
development. Major and fundamental reforms          as the semi-feudal production relations long
greatly increased TFP growth and are now            associated with Bihar’s agriculture virtually
promoting rapid gains in water-use efficiency       disappeared. The number of attached laborers
in Chinese agriculture. Total water use dropped     and landlords has declined significantly, while
sharply with diversification into less water-       casual landless agricultural labor and numbers
intensive crops and better technology. The          of poor peasants have increased. High and
greater openness of Brazil, Indonesia, and          growing population pressure on limited land
China has also paid significant dividends in        has also led to an increase in the proportion of
production efficiencies and diversification, as     nonagricultural households and migration.
well as better access to technology, whether
from international public research systems,         The challenges of structural transformation are
other national public research organizations,       more acute in Bihar relative to India as a whole.
or the private sector (notably in relation to       Agriculture’s share of GSDP is declining rapidly
genetically modified crops). More efficient use     (from 43 percent in 1980–81 to 18 percent in
of scarce resources is the only way for India to    2009–10). The reduction in agricultural labor
sustainably raise land and labor productivity.      has been much slower, resulting in low labor
Collectively, BIIC can benefit from technological   productivity. These developments do not imply
collaboration in developing and adapting            stagnation, however. Land productivity has
environmentally friendly and climate-smart          changed considerably: Average yields rose at
technologies for sustainable agricultural           2.5 percent per year for rice and 2.3 percent per
growth.                                             year for wheat.

                                                    Within the state, diversification has distinct
Change at the Micro Level: Insights                 regional patterns. For example, cereals continue
from Village Studies                                to dominate in some districts (as in Rohtas),
Perspectives on structural change at the micro      and noncereal production is more prevalent in
level provide insights to complement the            others (as in Gaya, Nalanda, and Purnia). The
macro-level analysis by pointing to the likely      spread of new technology also shows significant
impacts of policies and institutions on farmers’    regional differences.
productivity and incomes. This section presents
findings from a longitudinal study examining        Another aspect of structural transformation
structural changes at the village level.            is the significant diversification in men’s
                                                    occupations (Table 7). Only 27 percent of income
                                                    now comes from agriculture (either from own
Longitudinal analysis of change in 36
                                                    production or wage work in agriculture), and
villages of Bihar
                                                    only 37 percent of males classify agriculture as
A longitudinal study by Rodgers and Sharma          their primary occupation. Levels of migration
(2011) spanned the period from 1981–83 to           out of the village are high. Migration income



  50
                                                                                  Chapter 5: Structural Change




Table 7:  Distribution of income by source and caste, 2011 (%)
                                                       Casual
             Own        Wage        All     Nonagric. wage in      Regular
Caste/    agricultural work in agricultural   own     non-agric. employment Other
community production agriculture income production sector          income   income Remittances
Brahmin/      32.7       0.2       33.0       10.5      0.5          13.1     23.0    20.0
Kayastha
Bhumihar/     24.8       0.1       24.9        6.1      0.3          19.6     13.7    35.4
Kshatriya
Kurmi         19.0       1.9       20.8       12.8      1.9          27.3     25.1    12.0
Yadav         45.4       2.2       47.6        8.4      2.4          10.0     14.0    17.6
Koeri         13.1       0.6       13.7       50.3      0.2           3.1      6.2    26.5
OBC II        14.2       1.3       15.5       32.0      9.9          14.4     10.9    17.4
OBC I         20.5       7.7       28.3        9.7     10.1           7.0     11.8    33.1
SC/ST         11.8       9.2       21.0        4.3     18.0           6.9     16.7    33.0
Muslim        18.8       3.9       22.7        4.4     15.8           7.6     16.9    32.5
Total         22.3       4.2       26.5      10.7       8.2          10.6    16.0     28.0
Source:  Rodgers and Sharma 2013.
Note:  SC/ST = scheduled castes and tribes. OBC = Other backward castes.


is the largest share of household income                      all economic classes (Annex 5, Figure A.5.1).
(at 28 percent, just above agriculture). At                   Even agricultural laborers derive only
least initially, migration seems to have been                 25 percent of their income from agriculture
a response to the lack of opportunity in                      and 34 percent from remittances. Poor peasants
local labor markets. Migration affects rural                  have similar sources of income, but they largely
production systems by pushing up local wages,                 undertake nonagricultural production activities.
promoting labor-saving cultivation techniques,                The middle and big peasants have a higher
and increasing the feminization of agricultural               share of agricultural income. Even so, it is still
labor markets.                                                less than half their income.

Income sources vary by caste (Table 7). The
                                                              Implications of Observations at the
highest share of income comes from agriculture
for Yadav and Brahmin/Kayastha castes; it comes               Micro Level
from wage or other income for scheduled castes                This study reveals that even villages in
and tribes (SC/ST), and from nonagricultural                  poor and backward areas are experiencing
production activities for Koeri and other                     significant changes. There is evidence of
backward castes (OBC) II. Remittances are                     rising real incomes, increasing agricultural
important for all, but especially for Bhumihar/               productivity, and significant benefits from
Kshatriya, OBC I, SC/ST, and Muslims.                         agricultural diversification. Structural
                                                              transformation has started, but a substantial
Nonagricultural income, dominated by                          share of the population remains in agriculture,
remittances, is the main source of income for                 with falling labor productivity. Bihar is



                                                                                                           51
Republic of India: Accelerating Agricultural Productivity Growth




somewhat unusual in the role that migration        These findings provide unique granularity to the
has played in its development. Migration has       macro perspective on structural transformation
paid dividends, but Rodgers and Sharma (2011)      at the national and international level. They
raise concerns about its sustainability as an      highlight the complexity of the development
engine of continued growth. Nevertheless, the      challenge and indicate the need for more
person-land ratio is worsening, calling for more   customized analysis to identify strategic
rapid nonfarm and on-farm diversification of       priorities at the local level, an issue addressed
incomes.                                           later in Section E of this report.




  52
C.	 Sources of Agricultural Growth




A
        s labor exits agriculture, increasing       productivity is Total Factor Productivity (TFP),
        the productivity of the labor force         which estimates the returns to the total input
        that stays behind becomes critical to       mix used in production. Multiple sources of
sustain inclusive growth and food security,         data are used at different levels of aggregation
with a progressively smaller agricultural labor     to put together a comprehensive picture of
force feeding a correspondingly growing             productivity in Indian agriculture.
nonagricultural population. When labor
is leaving agriculture more slowly than             The analyses are conducted at the (i) national
desirable—as it is in India—increasing output       level for the sector using data going back to
becomes an even more important means                the 1960s; (ii) at the state level using data
of increasing labor productivity. Chapter 2         going back to the 1980s; (iii) at the individual
established the critical role of weather—           crop level using the aggregate state-level Cost
sustained negative rainfall anomalies—in the        of Cultivation data, derived from extensive
slowdown in agricultural growth from the mid-       annual surveys undertaken by the Ministry
1990s to mid-2000s. A key outstanding question      of Agriculture; and (iv) at the household
related to the subsequent recovery in growth        level—both at the whole farm and individual
is whether the drivers of growth changed            crop level—using the NCAER-REDS database,
qualitatively in the mid-2000s. Answering this      which has unique panel data on over 4,000
question is important for assessing whether the     households from across 17 states for the
current growth episode can be sustained.            years 1982, 1999, and 2007. These data help
                                                    triangulate the main findings at different
This section uses a number of analytical tools      levels, using different data sources and
and methodologies to dig deeper into the            using different methodologies to provide a
sources of agricultural productivity growth and     comprehensive assessment from different
identify key constraints. Productivity is broadly   perspectives. The picture that emerges is thus
defined as the total (real) value of output per     a robust assessment of the current status of
unit of input. A comprehensive measure of           Indian agricultural productivity.




                                                                                                 53
6                      Growth Decomposition




Changes in Subsector Contributions                                         Contribution of subsectors to growth
                                                                Figure 36: 
                                                                           in agricultural GDP
The Indian National Accounts define the
primary sector to include agriculture (crops and                           100%
                                                                           80%
livestock), forestry and logging, and fisheries in
                                                                           60%
the category of Agriculture and Allied Activities
(AAA), on which aggregate statistics are reported.                 Share   40%
                                                                           20%
AAA GDP grew at an annual average of 3 percent
                                                                            0%
(in 2004/05 constant rupees) between 1980/81                               -20%
and 2011/12. Agriculture more narrowly defined                                    1950-59 1960-69 1970-79 1980-89 1990-99 2000-07
                                                                 Fish              4.46    5.92    4.03    5.92    6.44    4.02
as crops and livestock grew at an almost                         Forest            0.88    15.98   -2.3    -0.54   1.13    1.43
identical rate.67 Total GDP has grown much                       Livestock         11.04   4.45    34.36 37.03     27.26 29.12
                                                                 Crop Sector       83.62   73.64   63.9    5758    65.17 65.43
faster, resulting in a rapid decline in agriculture’s
share from 52 percent in 1950/51 to 37 percent                  Source: 
                                                                        National Accounts Statistics, CSO, in Kumar and
in 1980/81 and 14.5 percent in 2011/12.68 The                           Jain 2012.
structure of subsectoral growth has changed
relatively little over time (Figure 36).                        Figure 37: Shares of crops in sectoral growth
                                                                100%
Within the crops sector, growth is increasingly                                                                          Others
                                                                 80%                                                     Scane
driven by growth in high-value commodities                                                                               Vegs
other than food grains, but the overall structure                60%                                                     Fruits
of the sector in terms of area and value shares                                                                          Sp&Cond
has changed relatively slowly (Annex 6,                          40%                                                     Pltn crps
                                                                                                                         Fibres
Table A.6.1). The contribution of high-value                                                                             Oilseeds
                                                                 20%
fruits and vegetables to overall crop growth                                                                             Pules
rose dramatically, from 22 percent in the                         0%                                                     Cereals
1980s to 54 percent in the 2000s (Birthal et al.                              1980s        1990s        2000s
                                                                -20%
2014) (Figure 37). Fibers also increased their
contribution.                                                   Source: Birthal et al. 2014.

67	 Between 1950–51 and 2011–12, the annual agricultural
    growth rate is estimated to be 2.6 percent.                 The current structure of the crops sector
68	 Total GDP growth rate was on average 5.9 percent per year
    between 1980–81 and 20011–12. GDP growth in the 2000s
                                                                suggests a misallocation of resources, as
    averaged 7.6 percent per year.                              reflected in the divergence of the share of


                                                                                                                            55
Republic of India: Accelerating Agricultural Productivity Growth




            hare of value divided by share of area
Figure 38: S                                                                                Table 8: Growth in crop yields
           by crop type
                                                                                             Crops/crop groups               1980s       1990s       2000s
8                                                                                            Rice                             3.15        1.21         1.42
           1980s         1990s      2000s
7
                                                                                             Wheat                            3.24        1.82         0.73
6
                                                                                             Maize                            2.04        2.22         2.27
5
4                                                                                            Gram                             2.48        1.53         1.16
3                                                                                            Arhar                            0.07        0.13         0.94
2                                                                                            Groundnut                        1.74        1.34         1.76
1                                                                                            Rapeseed and mustard             3.00        0.38         2.13
0                                                                                            Soybeans                         5.27        1.91         1.71
                              Rape&Mus

                                          O. Oilseeds




                                                                     Pltn Crops
                                                                    Sp. & Cond.
                   Chickpea




                                Soybean




                                                        O. Fibres
    Wheat

    O. Cers
     Maize




                      P’pea

                                  G’nut




                                                          Cotton




                                                                          Fruits

                                                                                   S’cane
       Rice




                                                                           Vegs.
                                                                                             Cotton                           4.21       -1.40       10.29
                                                                                             Sugarcane                        0.21        0.79         0.59
                                                                                             Fruits                          -2.21        1.81        -1.48
Source: Authors, using data from Birthal et al. 2014.                                        Vegetables                      -2.46        0.38         1.31
                                                                                            Source:  Birthal et al. 2014
area allocated to individual crops relative
to their share in the total value of output.                                                but yields of most crops have stagnated, with
Traditional food grains and oilseeds contribute                                             the notable exception of cotton (Table 8).
proportionately less to value than the area they                                            Growth thus seems to be rooted in rising prices,
occupy (ratio less than 1), while the high-value                                            rather than area or yields.
crops contribute significantly more (Figure 38).
High-value crops thus account for an increasing
share of the value of agricultural output, but                                              Components Driving the Value
they do not occupy a correspondingly large                                                  of Crop Production
share of cropped area. Food grains occupy about
65 percent of the cropped area and contribute                                               Given the changes and uneven performance
41 percent to the value of production, whereas                                              in agriculture over the past three decades
horticultural crops contribute 27 percent to the                                            (Chapter 2), it is important to analyze sources
value and occupy 7 percent of cropped area.                                                 and patterns of growth. Using data from
This pattern may reflect structural constraints                                             1980–81 to 2009–10, Birthal et al. (2014)
or other distortions facing different crops.                                                decompose the value of total crop production
                                                                                            into contributions by area, yield, price, and
Several crops saw impressive growth in the                                                  diversification.70 The data and methodology
value of output.69 Area reallocation, especially                                            are described in Annex 6. The 20 major
starting from a low base, explains some growth,                                             agricultural states used in the analysis are
                                                                                            clustered into four regions—North, East,

69	 High growth was recorded in cotton (10.7 percent annual                                 70	 Increases in the gross value of output can be due to price
    growth in the 2000s) and soybeans (30 percent growth in                                     increases of individual crops, increased output of each
    the 1980s, with continued growth of about 9 percent into                                    crop, or a shift in the crop mix toward higher-value crops.
    the 2000s). Maize, chickpeas, and other oilseeds experienced                                Output is a product of area and yield. The fifth component
    growth at or above 5 percent in the 2000s. Plantation crops,                                (interaction; see Annex 6), represents the residual or
    fruits, and vegetables also had high growth rates in the 1990s                              unexplained change, but it is empirically insignificant so is
    and 2000s.                                                                                  not discussed here.




    56
                                                                                                                                                                                                                  Chapter 6: Growth Decomposition




Figure 39: Decomposition of growth in the crops sector
             150




             100




              50
  Millions




               0
                    1981-82
                              1982-83
                                        1983-84
                                                  1984-85
                                                            1985-86
                                                                      1986-87
                                                                                1987-88
                                                                                          1988-89
                                                                                                    1989-90
                                                                                                              1990-91
                                                                                                                        1991-92
                                                                                                                                  1992-93
                                                                                                                                            1993-94
                                                                                                                                                      1994-95
                                                                                                                                                                1995-96
                                                                                                                                                                          1996-97
                                                                                                                                                                                    1997-98
                                                                                                                                                                                              1998-99
                                                                                                                                                                                                        1999-00
                                                                                                                                                                                                                  2000-01
                                                                                                                                                                                                                            2001-02
                                                                                                                                                                                                                                      2002-03
                                                                                                                                                                                                                                                2003-04
                                                                                                                                                                                                                                                          2004-05
                                                                                                                                                                                                                                                                    2005-06
                                                                                                                                                                                                                                                                              2006-07
                                                                                                                                                                                                                                                                                        2007-08
                                                                                                                                                                                                                                                                                                  2008-09
                                                                                                                                                                                                                                                                                                            2009-10
              -50


                                                                      Area                                Yield                                Price                                Diversiﬁcation                                              Interaction

             -100

Source: Birthal et al. 2014.


West, and South, described in Annex 6 – and                                                                                                               made a consistent, moderate contribution to
regional decompositions are done to account                                                                                                               growth throughout the study period—about
for the effects of heterogeneity (for example,                                                                                                            one-quarter of total growth, somewhat less than
in resources, climate, infrastructure, and                                                                                                                might be expected from a rapidly transforming
institutions) on crop choice and outcomes.71                                                                                                              production structure. This picture is consistent
                                                                                                                                                          with the observation that the broad structure
Clear dynamics emerge at the national level,                                                                                                              of agriculture has changed relatively little at
along with worrying recent trends. The                                                                                                                    the national level. Shares of cereals and pulses
dynamics of agricultural growth through the                                                                                                               in output have declined somewhat (from
distinct growth periods emerge clearly in the                                                                                                             42 percent in the 1980s to 37 percent in the
decomposition of growth (Figure 39). Yields                                                                                                               2000s), and shares of horticultural crops have
dominated growth until the mid-1990s, as green                                                                                                            risen (from about 20 percent to 28 percent).72
revolution technology spread, supported by                                                                                                                Rice and wheat maintained their shares in
policies and investments promoting irrigation,                                                                                                            production. The reduction in cereals was almost
fertilizer, and HYVs. Diversification has also                                                                                                            entirely in coarse cereals and pulses.

71	 The East includes Assam, Bihar, Jharkhand, Odisha, and West                                                                                           The contribution of prices to growth increased
    Bengal; North includes Haryana, Himachal Pradesh, Punjab,                                                                                             through the 1990s. Prices were the dominant
    Uttarakhand, and Uttar Pradesh; West includes Chhattisgarh,
    Gujarat, Madhya Pradesh, Maharashtra, and Rajasthan; and
    South consists of Andhra Pradesh, Karnataka, Kerala, and                                                                                              72	 Soybeans and cotton gained shares over time but remain a
    Tamil Nadu. Data on other states were not available.                                                                                                      relatively small part of aggregate output.




                                                                                                                                                                                                                                                                                                            57
Republic of India: Accelerating Agricultural Productivity Growth




source of growth in the mid-1990s, coinciding                        The two worrying trends are the declining
with significant increases in the MSPs for rice                      contribution of yields and the emergence of
and wheat and favorable terms of trade for                           prices as the main driver of growth toward the
agriculture.                                                         end of the 2000s. In 2010, 55 percent of the
                                                                     increase in the real value of output resulted
As seen in Chapter 2, growth slowed sharply                          from price increases. Yield and diversification
between 1996-97 and 2002-03, as real prices                          (to high-value crops) made relatively small
declined, yields stagnated, area declined,73                         contributions, raising serious concerns about
and adverse rainfall anomalies beset                                 the sustainability of growth.
agriculture. Throughout that turbulent period,
diversification proved to be a steady source of                      The four regions of the country are
growth in absolute terms; in relative terms,                         experiencing agricultural change in different
diversification was the predominant source of                        ways. Cereals still dominate the area under
growth.                                                              production in the North and East, but they
                                                                     now share the lead with horticulture in
During the 2000s, the average contribution                           terms of value. The West and South are more
of the four factors—area, yield, price, and                          diversified. The West has benefited from the
diversification—appears balanced, but in fact                        new cotton and soybean technology, and the
important shifts occurred within the decade.                         South from horticulture and spices. The main
Area and yields propelled growth early in                            driver of growth in the North—through the
the recovery phase, as area came back into                           late 2000s—was yield (Figure 40), and the
production and yields rebounded with normal                          continued predominance of rice and wheat
rainfall. Real prices continued to decline for                       reflects the impact of research on those
another couple of years, slowing recovery                            crops. Prices have become an increasingly
somewhat.74 Since 2007, as one would                                 important driver of growth in the East, with
expect, area expansion has slowed.                                   important contributions also from yields
Diversification, in absolute and relative terms,                     and diversification. The four components of
remains a consistent but modest source of                            growth are more balanced in the South and
growth.75                                                            West, but diversification has been the major
                                                                     driver in both regions. In the South, as surface
                                                                     irrigation systems recovered from prolonged
73	 To facilitate analysis, area, production, and price series are   rainfall deficits, area made a fair contribution to
    smoothed using the HP filter, so the trends appear smoother
    than if raw annual data had been used. The output trend
                                                                     growth.
    thus approximates the decadal growth trend discussed in
    Chapter 2.
74	 Trends in the components of WPI inflation indicate that
    real prices for agricultural commodities—the WPI index for
                                                                     Summary of Findings
    agricultural commodities, divided by the total WPI index—
    fell broadly, with the exception of fruits and vegetables, on    Recent growth had little to do with sustainable
    account of a faster rise in nonagricultural WPI components.      sources such as yield or productivity. Despite
75	 For the past four decades, net area under cultivation has
    remained steady at about 140 million hectares (not exceeding     diversification in the value of output, the
    143 million hectares) (Ministry of Agriculture Land Use          national and regional analyses find that the
    Statistics). The slow expansion of irrigation has meant that
    irrigation intensity and hence gross cultivated area have not
                                                                     production structure has changed little. Note
    expanded much since 2004/05.                                     the important regional variations: The South



   58
                                                                                                                    Chapter 6: Growth Decomposition




Figure 40: Sources of agricultural growth by region
                                                     North                                                                            East
                   20                                                                                  20


                   10                                                                                  10
        Millions




                                                                                           Millions
                    0                                                                                   0
                            1980/
                         81-88/89



                                             1988/
                                          89-96/97



                                                            1996/
                                                         97-04/05



                                                                             2004/
                                                                          05-09/10




                                                                                                                1980/
                                                                                                             81-88/89



                                                                                                                              1988/
                                                                                                                           89-96/97



                                                                                                                                                1996/
                                                                                                                                             97-04/05



                                                                                                                                                              2004/
                                                                                                                                                           05-09/10
                   -10                                                                                 -10



                               Area          Yield                     Price                                       Area          Yield                   Price
                               Diversiﬁcation                    Interaction                                       Diversiﬁcation                  Interaction




                   60                                South                                             40                             West


                   40
                                                                                                       20
    Millions




                                                                                            Millions
                   20
                                                                                                        0
                    0
                              1980/
                           81-88/89



                                               1988/
                                            89-96/97



                                                                1996/
                                                             97-04/05



                                                                               2004/
                                                                            05-09/10




                                                                                                                   1980/
                                                                                                                81-88/89



                                                                                                                                 1988/
                                                                                                                              89-96/97



                                                                                                                                                   1996/
                                                                                                                                                97-04/05



                                                                                                                                                                 2004/
                                                                                                                                                              05-09/10
                   -20                                                                                 -20




                                    Area          Yield                   Price                                     Area          Yield                   Price
                                    Diversiﬁcation                  Interaction                                     Diversiﬁcation                  Interaction

Source: Birthal et al. 2014.


and West have diversified, whereas the North                                           term, was mainly spurred by real price changes
remains largely invested in traditional cereal                                         rather than growth in yields or productivity.
crops, and in terms of value, production in                                            This worrisome development indicates
the East is now dominated by cereals and                                               that growth is not underpinned by a more
horticultural crops.                                                                   sustainable source—productivity. The next
                                                                                       chapter looks closely at how productivity has
A cause for concern is that the recent growth                                          evolved, especially in recent years, across India’s
spurt, while good for farmers in the short                                             heterogeneous agricultural landscape.




                                                                                                                                                                         59
7                       Total Factor Productivity
                        Growth




I
    ndia’s binding land constraint means that                       important to raise agricultural incomes and give
    agricultural growth depends crucially                           farmers an incentive to invest in agriculture.
    on making land more productive. Land                            But rising prices—especially if managed by
productivity is often equated with yields—                          policy—may mask growing inefficiencies by
generically referring to output per hectare                         boosting farm profits even as yields decline
for crops and per animal for livestock.76 Yet                       (Rao and Dev 2010). The critical question
for some time analysts have warned about                            is whether growth is underpinned by
declining productivity, based on worrisome                          improvements in productivity in response
trends in rice and wheat performance.77 Despite                     to the improved incentives, or if growth is
its position in the vanguard of the green                           the result of higher input use (perhaps made
revolution, over time India has performed less                      further financially enticing through continued
impressively than other major agricultural                          subsidies). The risks to overall economic
producers (Chapter 3), and evidence has                             growth, environmental sustainability, and food
emerged that soil and water degradation may                         security look very different depending on the
now limit productivity in the pioneering green                      answer to that question.
revolution states.78
                                                                    Agriculture has rebounded since 2004/05. In
Previous chapters presented evidence that                           a context of increasingly binding resource
agricultural growth is being driven less by                         constraints and the uncertainties posed by
sustainable improvements in yields than by                          climate change, it becomes important to
rising real prices and possibly unsustainable                       understand how productivity has evolved in
input use. To be clear, rising real prices are                      recent years to identify potential policy and
                                                                    strategic issues. This chapter seeks to shed light
76	 As noted, since the 1960s the government’s agricultural         on the parallel issues of “technology fatigue”
    policies have focused on intensification, with particular
    attention to rice and wheat yields. Agricultural technology     and “policy fatigue” to help guide strategy
    and inputs have been actively promoted in Indian agriculture    and policy in support of sustainable growth in
    since the 1950s and were the basic ingredients of the green
    revolution—HYVs, fertilizer and other agro-chemicals,           agricultural productivity. It does so by analyzing
    irrigation, and mechanization. The progressive impact of this   Total Factor Productivity (TFP) over space and
    effort through the 1970s and 1980s is visible in agricultural
    growth and aggregate GDP growth (Bosworth, Collins, and         time, building on past work by updating and
    Virmani 2007).                                                  deepening the analysis, and seeking to reconcile
77	 See, for example, Kumar and Mittal (2006); Chand, Kumar
    and Kumar (2011).
                                                                    divergent findings from different studies using
78	 See, for example, Murgai, Ali, and Byerlee (2001).              existing secondary data. The next chapter



                                                                                                                 61
Republic of India: Accelerating Agricultural Productivity Growth




provides complementary evidence from micro-                     “accounting” for the contribution of all factor
level data to add a fresh perspective.                          inputs to the growth in output. For growth to
                                                                be sustainable in both ecological and economic
Earlier studies of TFP in Indian agriculture                    terms, it is critical to improve TFP (Ehui and
have yielded mixed empirical results (Annex                     Spencer 1993, Cassman and Pingali 1995).
7 summarizes the literature). Most of these                     Ecologically, improved TFP ensures that critical
studies cover periods prior to the recovery                     natural resources such as land and water are
in 2004/05. To build a convincing body                          used effectively; economically, it reflects income
of evidence, the analysis relies on results                     growth.
from complementary ongoing independent
studies, updates of recent insightful studies
when possible, and new background studies                       Drivers of Productivity Growth at
conducted to fill some of the knowledge gaps.                   the National Level
What sets the new studies apart is their use
                                                                Recent estimates (and updates) of TFP for Indian
of more recent data (vetted for reliability);
                                                                agriculture provide important insights into
an effort to triangulate and deepen the
                                                                the nature and drivers of productivity growth.
understanding of key issues through analyses
                                                                Depending on the methodology and level of
at different levels of disaggregation; the use
                                                                aggregation, however, results vary substantially
of alternative methodologies that allow more
                                                                across studies, so careful scrutiny is required
insight into the main drivers of productivity;
                                                                to reconcile and interpret the findings. This
and an assessment of changes in productivity at
                                                                exercise reveals that the differences primarily
the micro/farm level based on panel data.
                                                                result from the level of aggregation used in each
                                                                study and also depend somewhat on the level of
The methodology and data are detailed in
                                                                accounting for different inputs.
Annex 7. Very briefly, yield improvement—
broadly defined to refer to land productivity—
                                                                At the highest level of aggregation—the sector-
can be achieved by using more inputs other
                                                                wide level, including all crops and livestock
than land (intensification), using better inputs
                                                                products—there are four estimates of TFP,
(new technology), and/or using inputs more
                                                                based on the method and data used. Given
effectively (efficiency).79 Understanding how
                                                                the changing structure of Indian agriculture,
each of these drivers contributes to productivity
                                                                where horticultural crops and livestock have
gains is important. In traditional analyses
                                                                increasing shares in aggregate output but the
of sources of growth, the contributions of
                                                                relatively lower-value crops continue to absorb
technology and efficiency are referred to
                                                                more resources (their historically large share
jointly as Total Factor Productivity (TFP).
                                                                of public expenditures, considerable policy
TFP is typically measured as a residual after
                                                                support, and natural resources such as land and
                                                                water), an alternative series is also estimated.
79	 “Intensification” implies moving along the production       This alternative series is restricted to crops that
    surface with a given technology, “new technology” is an
    outward shift of the production surface, and “efficiency”   have traditionally been considered important.
    measures the distance between the current level of output   Not by coincidence, these are the major crops
    and the optimal level that could be achieved, as defined
    by the current level of technology (or a move toward the
                                                                that are grown in the main agricultural states
    production surface).                                        and covered in the Ministry of Agriculture’s



   62
                                                                            Chapter 7: Total Factor Productivity Growth




annual Cost of Cultivation (CoC) surveys.                                      TFP estimates at the national, sector-
                                                                     Table 9:  
These CoC data provide a comprehensive cost                                    wide level
breakdown by input and permit a much better                          Period           DEA-FAO       GA-FAO GA-GDP            GA-NA
accounting of the inputs used in production.                         1980–2008          1.3          1.4    1.5               1.9
The state-level input-output estimates provided                      1980–1997          1.6          1.6    1.6               2.2
by the CoC database are based on large annual                        1997–2003          0.9          0.4    0.9               0.6
surveys and form the basis for estimating the
                                                                     2003–2008          2.2          3.2    1.8               5.3
costs of cultivation to establish the official
                                                                     2003–2009          1.7          2.5      –                 –
Minimum Support Prices for certain crops
                                                                               uthors, using data from Fuglie 2012; updated
                                                                     Source:  A
(MSPs).80 For those reasons, the data provide
                                                                              Bosworth and Collins database (World Bank
a much better and a more comprehensive                                        2012a); Rada 2013; and Nin-Pratt 2013 (personal
basis to “account” for inputs when deriving                                   communication).
TFP estimates. The data also make it possible
to calculate reliable TFP estimates at the state                     derived from FAO data (from a comprehensive
level, which can be aggregated to the national                       and internationally comparable database put
level as a weighted average using area shares in                     together by Fuglie, 2012). The DEA-FAO refers
total.                                                               to the nonparametric Data Envelopment
                                                                     Analysis (DEA) methodology applied to the same
An important difference between the CoC                              database, which allows a “clean” comparison
data and the National Accounts data—and the                          of the two methodologies without concerns
main difference between the TFP estimates                            about differences in data. The GA-GDP also
that follow for all of agriculture versus the                        uses the growth accounting technique but a
“traditional” or major crops—is the exclusion of                     slightly different accounting procedure and
high-value agricultural sectors, namely livestock                    agricultural GDP (valued added for crops and
and horticultural crops (with the exception of                       livestock) data. The GA-NA is an application of
potatoes and onions for major producing states)                      growth accounting using more detailed national
in the CoC data. The estimates using only the                        accounts data for the value of output (for all
traditional crops provide a useful contrast with                     crops and livestock) combined with secondary
the performance of the full sector.                                  data on factor inputs from different official
                                                                     national sources. In sum, the main differences
                                                                     in the three GA series is their use of different
TFP estimates for all crops from four                                data and their method of accounting for inputs
sector-wide studies                                                  (discussed in Annex 7). The period of analysis is
Table 9 summarizes the main results from the                         from 1980 onward, when the green revolution
four sector-wide studies. The GA-FAO refers to                       was well established. The analysis focuses
the growth accounting method using a gross                           on the distinct growth episodes described in
output index covering all crops and livestock                        previous chapters.81

                                                                     Major conclusion and substantive findings.
80	 The detailed survey data are not available to the public. Only
    the processed data, including yields and costs of cultivation
                                                                     A major conclusion from the findings is
    by individual inputs on a per hectare basis, are available.
    The cost elements covered are comprehensive, however,            81	 The slowdown between 1997 and 2005 “bottomed” out in
    and include imputed costs of family labor, land, and other           2003, the year with the large negative shock to agricultural
    variable costs such as cost of capital.                              production and productivity since 1980.




                                                                                                                                 63
Republic of India: Accelerating Agricultural Productivity Growth




that over the long run (from 1980 to 2008               provides some confidence that the specific
or 2009, depending on the study) different              insights provided by each methodology are
methodologies and models delineate very                 useful for characterizing the underlying
similar trends (see Annex 7, Figure A.7.1),             dynamics of productivity growth.
allaying concerns that different
methodologies may give different results (Coelli        The GA-FAO estimates (Figure 41), using
and Rao 2006; Fuglie 2012; Lele et al. 2013).           data from 1961 onward, show progressive
In fact, the biggest difference arises between          improvement in TFP over the decades, except
estimates that use the same method but                  for the 1990s, with their weather anomalies.
different data.                                         Among individual inputs, irrigation was an
                                                        important driver of growth in the earlier
Two substantive findings emerge from                    decades; irrigation expanded much more
Table 9. First, the TFP growth estimate varies          slowly in the 2000s, suggesting that expansion
significantly depending on the choice of years,         may be reaching its limits. Growth in inputs
highlighting the role of the exogenous weather shocks   other than land and irrigation has been an
identified earlier. The start and end dates make        important source of growth, but most of the
a significant difference in the growth rate.            recent growth in output is due to growth
For example, 2009 was a severe drought year,            in TFP.
whereas the previous three years experienced
good rainfall, so changing the terminal period          Using different growth accounts, the Bosworth-
by just one year makes a big difference in              Collins model (Bosworth and Collins 2003)
the TFP estimate, as seen in the last row of            provides a different perspective on the basic
Table 9.                                                sources of growth (Table 10). It shows the
                                                        limited role of education in agricultural
Second, the TFP estimates for the post-2003 period      productivity over time along with two
show strong growth, but this finding needs to be        encouraging emerging trends. First, the pace at
treated with guarded optimism. Previous studies
noted decelerating TFP trends after the mid-                       Contribution of inputs and TFP to
                                                        Figure 41: 
1990s, but with the benefit of a longer time                       output growth
horizon (and hindsight), it is clear that the           4.0%
decline was associated with negative rainfall           3.5%                                 1.3%            2.1%
shocks. Since 2003, TFP has started to recover,         3.0%     1.2%                1.0%
                                                                                                     1.1%
but this finding needs to be tested with                2.5%
additional years’ data to determine whether it is       2.0%                  0.5%
robust or reflects a recovery from the previous         1.5%
slowdown.                                               1.0%
                                                        0.5%
Insights from each methodology. The                     0.0%
comparability of estimates derived from these                   1960-09     1961-70 1971-80 1981-90 1991-00 2001-09
                                                        -0.5%
varied methodologies and data sources provides
                                                                  Irrig   New Land          Input/Area         TFP
confidence that the evidence of recovery is
strong. This consistency in the estimates also          Source: Calculated from Fuglie database, 2011.




  64
                                                                     Chapter 7: Total Factor Productivity Growth




           Agricultural sector growth accounts by growth episode, 1980–2008 (trend growth rates)
Table 10:  

                 Output                        Output/                     Output/worker—contribution of:
Period           Growth     Employment         worker            Capital       Education          Land              TFP
1980–1997          3.2            1.2            1.9               0.2            0.2             -0.2              1.6
1997–2003          2.3            0.9            1.3               0.6            0.3             -0.6              0.9
2003–2008          3.5            -0.2           3.7               1.2            0.2              0.5              1.8
Source:  Updated Bosworth and Collins data (World Bank 2012a).


which labor is shifting out of agriculture and              a result of technological progress.82 Efficiency
into more productive sectors appears to have                change measures a movement toward the
picked up since 2003, with implications for                 production frontier, or a closing of the gap
the structural transformation in the economy.               between the current level of production
Another trend—the lack of capital deepening                 and the frontier—which is the level that
in agriculture—also appears to be changing,                 is technically feasible with the same set of
with significantly faster growth in capital                 inputs (“catching up”). Efficiency change is
stock between 2003 and 2008 relative to earlier             a joint outcome of several factors, including
periods. With land remaining constant over                  knowledge about available technology, the
time, these trends translate into rising labor              proper use of inputs, the incentives that
productivity, which shows growth rates in the               influence farmers’ choice of technology and
post-2003 period to be almost double those in               inputs, access to markets (for inputs and
earlier periods. As noted, these results must be            outputs), and other policies and institutions
viewed with guarded optimism, as the period of              affecting input-use decisions.
analysis after 2003 is relatively short, but they
do suggest that growth dynamics in agriculture              The DEA results are presented in Figure 42,
may be changing.                                            which shows trends in the indices for TFP,
                                                            technical change, and efficiency growth.
An important advantage of the DEA                           Table 11 summarizes the growth rates by time
methodology (other than the implied                         period. Four important conclusions emerge
production structure or assumption of                       from these findings. First, technological
competitive markets implicit in the GA                      progress has been the primary, consistent
methodology) is that it makes it possible                   driver of productivity growth over the past
to disentangle two key components of                        three decades. Second, efficiency improved
TFP: technical change and efficiency. This                  in the 1980s but then plateaued in the early
decomposition helps to identify whether                     1990s, indicating that as the production
productivity has grown because of changes
                                                            82	 In the case of a single output or a homogenous group of
in technology or changes in efficiency. Each                    outputs, technical change is straightforward technological
measure reflects the influence of different                     progress—an outward shift in the production function. In
                                                                the context of a composite output (such as a sector-wide
factors with different policy implications.                     aggregate), technical change is an outward shift of the
                                                                observed production surface that could occur because of
                                                                pure technological progress (as for individual products) or a
Specifically, technical change indicates an                     change in the output-mix, with a shift to more “productive”
outward movement of the production frontier,                    products in terms of input use.




                                                                                                                          65
Republic of India: Accelerating Agricultural Productivity Growth




Figure 42: P
            roductivity change in agriculture:                                                                 TFP growth in traditional crops and
           Technical progress versus “catching                                                                  agriculture as a whole: Sobering results
           up” with the frontier
                                                                                                                As noted, analyses at the sector-wide level
160
150
                                                                                                                include the fast-growing livestock and
140                                                                                                             horticulture (fruits, vegetables, and spices/
130                                                                                                             condiments) subsectors, generally referred to
120                                                                                                             as high-value agriculture. Both subsectors are
110                                                                                                             important for inclusive growth. Both have been
100                                                                                                             major drivers of growth in agricultural value
 90
                                                                                                                added in recent years. Yet neither subsector has
 80
                                                                                                                commanded the attention of policy and public
      1980
             1982
                    1984
                           1986
                                   1988
                                          1990
                                                 1992
                                                        1994
                                                               1996
                                                                      1998
                                                                             2000
                                                                                    2002
                                                                                           2004
                                                                                                  2006
                                                                                                         2008
                                                                                                                expenditure—which translates to resource
              TFP                 Catch-up                Change in Production Frontier                         allocation priorities at the farm level—to
Source: Authors, using FAO data.
                                                                                                                the same extent as the politically important
                                                                                                                traditional crops (cereals, pulses, oilseeds,
                                                                                                                cotton, sugarcane). Together the traditional
Table 11:  Decomposing TFP growth                                                                               crops account for about 90 percent of the
                                                         Efficiency                   Technical                 area under cultivation (average 2000–10), and
Period                     TFP growth                     change                       change                   they have been the focus of policy, public
1980-2009                           1.3                        -0.4                          1.7                expenditures (including substantial input
1981-1997                           1.6                        0.4                           1.2                subsidies), and public services.
1997-2003                           0.9                        -1.1                          2.0
                                                                                                                These traditional crops contributed less than
2003-2009                           1.7                        -0.8                          2.5
                                                                                                                half of the sector’s growth in 2000s (see
Source:  Authors, using FAO data.                                                                               Figure 37). With an overwhelming majority
                                                                                                                of resources tied up in these crops, a natural
frontier continued to shift outwards,                                                                           question is how their productivity compares to
production kept pace but the “yield gap”                                                                        that of the sector overall. Using the CoC data
was not closed further—in other words, no                                                                       described earlier, Kumar, Gautam, and Joshi
change in efficiency occurred. Third, efficiency                                                                (2013) estimate TFP for the traditional crops.
declined in the period from 1997 to 2003,                                                                       The crops used for the estimation contribute
likely due to weather shocks, but positive TFP                                                                  about half of the value of crop production but
growth was maintained with strong movement                                                                      account for about 80 percent of the national
of the production frontier—which more than                                                                      area (Annex 7, Table A.7.1 lists the crops
compensated for falling efficiency. Fourth,                                                                     included by state).
an even stronger surge in the production
frontier has occurred since 2003, but efficiency                                                                The results from this alternative growth
levels have continued to decline, though not                                                                    accounting exercise are summarized in
as rapidly as between 1997 and 2003. The                                                                        Figures 43 and 44. Figure 43 compares trends
implication is that the gap between realized                                                                    in TFP growth at the sector-wide level and for
and realizable productivity has widened.                                                                        the traditional crops. The comparison is stark




  66
                                                                                                                                         Chapter 7: Total Factor Productivity Growth




             FP trends, sector-wide and for
 Figure 43: T                                                                                                                              Sources of growth in the traditional
                                                                                                                                Figure 44: 
            traditional crops                                                                                                              crops subsector

                       200             Growth rate                                                                               7.00%                                                 6.29%
                                       Sector - wide                          1.77%                                              6.00%
                       180
                                       Traditional Crops                      0.28%                                              5.00%                                                2.25%
                                                                                                                                                          3.69%
                       160
TFP Index 1981 = 100




                                                                                                                                 4.00%      2.83%
                                                                                                                                 3.00%           0.20%    0.79%
                       140
                                                                                                                                 2.00%                                                4.04%
                                                                                                                                           2.63%          2.90%           -1.44%
                       120                                                                                                       1.00%
                                                                                                                                                                          1.00%
                                                                                                                                 0.00%
                       100                                                                                                                                                 -2.44%
                                                                                                                                -1.00%
                        80                                                                                                      -2.00%
                                                                                                                                -3.00%
                        60                                                                                                                1981/82-       1981/82-         1997/98-   2002/03-
                             1981
                                    1983
                                           1985
                                                  1987
                                                         1989
                                                                1991
                                                                       1993
                                                                               1995
                                                                                      1997
                                                                                             1999
                                                                                                    2001
                                                                                                           2003
                                                                                                                  2005
                                                                                                                         2007
                                                                                                                                          2008/09        1997/98          2002/03    2008/09

                                            Traditional Crops                            Sector-wide                                                      Input     TFP     Output

 Source: Kumar, Gautam, and Joshi 2013 and Rada 2013.                                                                           Source: Kumar, Gautam, and Joshi 2013.


 and sobering. Long-term TFP growth (between                                                                                    In the slowdown period, traditional crops saw a
 1980 and 2008) is estimated at only 0.28 percent                                                                               more dramatic downturn: TFP declined rapidly,
 for traditional crops, much less than the rate of                                                                              at –2.4 percent per annum. The subsequent
 more than 1.77 percent estimated for the sector                                                                                rebound has been impressive. TFP has grown
 as a whole. Most of the growth in productivity                                                                                 at about 2.3 percent, but it bears repeating that
 over the long run has come from a change in                                                                                    this growth should be treated with cautious
 the mix of outputs—from the increased shares                                                                                   optimism, as it may reflect a reversion to the
 of high-value crops and livestock products.                                                                                    trend that prevailed prior to 1997. The major
 Productivity growth in the major crops appears                                                                                 share (64 percent) of output growth still comes
 to have stagnated.                                                                                                             from increased input use.

 Breaking down TFP performance for the                                                                                          The more pronounced periodic shifts in TFP
 traditional crops by growth period provides a                                                                                  for the traditional crops than for the entire
 more nuanced assessment, but one that shows a                                                                                  sector indicate that productivity in traditional
 now-familiar pattern (Figure 44). An outstanding                                                                               crops may be more susceptible to weather
 feature through all periods is that inputs have                                                                                shocks. This is consistent with the earlier
 been the primary drivers of growth in output.                                                                                  finding that the livestock sector experienced
 TFP growth was modest—significantly less                                                                                       only a mild deceleration during the 1997–2005
 than at the sector-wide level (about 0.8 percent                                                                               slowdown, likely cushioning the aggregate TFP
 versus 1.6 percent, even during the extended                                                                                   shock. The continued reliance on subsidized
 period of expansion following the green                                                                                        input use as the main driver of growth for
 revolution (1980–97). With three-quarters of                                                                                   traditional crops may be a cause for concern.
 the growth being driven by inputs, concerns                                                                                    Starting from a low base, an increase in input
 raised by previous studies about the quality and                                                                               use (intensification) in the lagging states where
 sustainability of agricultural growth were well                                                                                it has historically been very low may suggest
 warranted.                                                                                                                     more inclusive growth, with the expectation



                                                                                                                                                                                          67
Republic of India: Accelerating Agricultural Productivity Growth




that TFP will improve as farmers gain more         Substantial variation in TFP across states.
experience in using inputs. But the continued or   Table 12 shows the estimated TFP growth rates
accelerated use of imbalanced chemical inputs      from both studies. Over the long run, both TFP
and overexploitation of natural resources such     estimates are close for some states (Gujarat,
as groundwater in the states already heavily       Haryana, and Tamil Nadu, for example) but
using these inputs would warrant urgent            show substantial differences for others (such
corrective action. Such analysis requires more     as Andhra Pradesh, Maharashtra, Odisha,
disaggregated analysis, accounting for the         Rajasthan, and West Bengal). The main reason
heterogeneity across the country, an issue that    for these differences (spatially as well as
is covered next.                                   temporally) is the scope of coverage of the crops
                                                   sector, specifically the exclusion of livestock and
                                                   high-value crops in the estimate for traditional
TFP at the State Level                             crops.
Agro-ecological as well as policy and
institutional diversity are hidden in aggregated   In recent years, the subsectors driving
analysis. Agriculture is a State Subject, so       productivity growth have been increasingly
performance may vary considerably depending        diversified. Even when livestock and high-value
on the prevailing governance environment.          horticultural crops are excluded, TFP growth is
Few previous studies have looked at                highly varied across crops and across states for
productivity at the disaggregated level. Chand     the same crop. Most states have improved their
et al. (2010) provide more detailed, crop-         overall performance significantly since 2003
specific TFP estimates for different states,       compared to the period before 1997 (ignoring
offering rare insight into a third level of        for the present the years of poor performance
disaggregation (the crop level). An important      falling between 1997 and 2003). Among those
finding from their study is that TFP growth        states, Bihar, Karnataka, Maharashtra, and
varies widely across crops and states. No          Odisha show remarkable improvement, while
previous studies have assessed TFP at the          Gujarat, Himachal Pradesh, and West Bengal
state level for all of the agricultural sector     are the outliers with negative growth rates. TFP
to determine how the states have performed         stagnated in Rajasthan and Uttar Pradesh.
in terms of productivity. Such an analysis is
important to identify broader sectoral policy      Different states have benefited from different
levers to promote growth.                          crops, the biggest gains being associated with
                                                   new technologies for cotton and maize. But the
Following the approach used at the aggregate       impact of the new technologies on productivity
national level, two new studies by Rada (2013)     growth has not been uniform across states, as
and Kumar, Gautam, and Joshi (2013) estimate       indicated by the highly varied performance
state-level TFP for 1980–2008 at the sector-       across states even for the same crops (shown for
wide and traditional crop levels, respectively.    selected crops in Annex 7, Figure A.7.2). Among
As before, the analysis covers about 54 percent    the major cereals, productivity in rice has
of the aggregate gross output from the crop        improved for several states, but it has slowed
sector and excludes the horticulture, spices/      considerably for wheat, except in reform-driven
condiments, and livestock subsectors.              Gujarat (Shah et al. 2009). New technologies for



  68
                                                                   Chapter 7: Total Factor Productivity Growth




           TFP estimates by state, traditional crops and sector-wide, 1998–2008
Table 12:  
                                                     Major crops                                Sector-wide
State                  1980–1997          1997–2003           2003–2008         1980–2008       1980–2008
AP                         0.9                0.5                  3.1             0.7              3.2
Assam                      0.6                -1.0                 2.0             0.3              1.5
Bihar                      0.3                -0.1                 5.1             0.2              1.3
Gujarat                    5.3                -2.2                 -5.4            2.3              1.8
Haryana                    2.6                -0.7                 4.6             1.6              1.4
HP                         0.7                -0.7                 -1.8            0.2              1.6
Karnataka                  1.0                -4.4                 3.7             0.4              2.1
Maharashtra                0.5                -0.7                 8.4             0.2              2.4
MP                         1.0                -5.7                 3.0             0.0              1.6
Odisha                    -2.8                -1.8                 4.2             -1.6             0.6
Punjab                     1.1                0.1                  3.2             0.9              1.6
Rajasthan                  0.8                -9.3                 0.1             -1.8             2.3
Tamil Nadu                 3.1                -3.2                 5.7             2.2              2.7
UP                        -0.8                -0.2                 1.1             0.7              1.4
West Bengal                1.4                -3.0                 -1.2            0.2              2.7
Kerala                                                                                              3.3
Source:  Kumar, Gautam, and Joshi 2013; Rada 2013.


maize and oilseeds benefited only some states.             Bt cotton technology has been a major driver of
Notably, the significant success of Madhya                 cotton productivity, making a huge impact very
Pradesh in increasing soybean output (which                soon after its introduction in 2002. Sugarcane,
has been an important driver of agricultural               on the other hand, has been subject to more
growth in Madhya Pradesh) appears to have                  government interventions and significant
been through increased input use, as TFP                   subsidies, which appear to be driving up input
growth was weak for soybeans in Madhya                     use even as productivity declines.
Pradesh. On the other hand, TFP improved
significantly for sunflowers in Maharashtra and            Determinants of TFP at the state level.
Karnataka.                                                 Given the wide variation in TFP across states,
                                                           the obvious question is what explains that
A vast difference can be seen in the                       variation. Several factors are likely to influence
performance of cotton and sugarcane, two                   productivity, including research, extension,
major cash crops in a number of states. Cotton             human capital, and infrastructure development,
shows consistent gains across states, except               not to mention climatic factors. TFP subsumes
for Karnataka, but sugarcane shows consistent              a large number of unobserved factors, making
declines in TFP growth across all major                    it difficult to assess the role of important policy
sugarcane-growing states and for both periods              levers and potentially yielding a misleading
(before and after 1995). As is well known, new             interpretation with policy implications.



                                                                                                          69
Republic of India: Accelerating Agricultural Productivity Growth




For example, technology is well established                       factors indicates the presence of other possibly
as a key driver of productivity both globally                     important factors.85
as well as in India. But slowing growth in
agriculture, accompanied by slowing growth                        Land that is multicropped will have higher TFP—a
in TFP, particularly for major crops, has                         tautology. The estimates confirm this association,
generated debates on “technology fatigue” and                     highlighting the role of irrigation, as cropping
the effectiveness of agricultural research and                    intensity is almost directly correlated with
extension. On the other hand, especially in the                   irrigation. An increase of 1 percent in cropping
context of the growth spurt since 2003, it is not                 intensity is associated with 0.6–0.7 percent higher
clear if the main drivers are indeed technology                   TFP (depending on the specification used), which
or diversification.                                               is substantial but also indicates that irrigation
                                                                  expansion does not lead to a 1:1 increase in area.
An analysis of the determinants of TFP
(Kumar, Gautam, and Joshi 2013), using state-                     As discussed, productivity can change because
level TFP indices for traditional crops and                       of a shift from low- to high-productivity crops
controlling for some of the confounding factors,                  or through the introduction of new or better
provides important insights with significant                      technology. To capture the crop-mix effect,
policy implications.83 The main findings are                      crop diversification (index) is included as a
summarized here.84                                                control variable. There is almost a one-to-one
                                                                  correspondence between diversification and TFP
State-specific (fixed) effects explain nearly half of the         growth, clearly demonstrating that diversification
observed variation in TFP, clearly demonstrating                  is a powerful driver of productivity growth, even within
the important role of state-specific factors,                     the more restricted group of traditional crops.
including policies, institutions, governance, and
public investments. Annual shocks (annual fixed                   Beyond these “control” variables, technology
                                                                  is perhaps the most well-established driver of
effects) by themselves explain only a small part of the
                                                                  productivity growth, and its contribution has
variation but are nevertheless significant. Rainfall
                                                                  not diminished over time. Two dimensions of
has the expected large impact. Together,
                                                                  technology that are operationally relevant are
these “control” factors explain over half of
                                                                  captured through “stock” variables to proxy
the observed variation in TFP. The inclusion
                                                                  technology generation (agricultural research
of policy variables captures a substantial
                                                                  output) and technology/knowledge dissemination
part of the state effects, but their continued
                                                                  (the impact of agricultural extension).86 Research
significance over and above the included
                                                                  85	 All regressions included state and annual dummy variables
                                                                      to capture the two dimensions of fixed effects and help focus
83	 A similar analysis using the broader, sector-wide TFP             on key policy variables. All regressions control for state
    estimates of Rada (2013) has not been undertaken.                 clustering effects to get robust standard errors for inference.
84	 See Kumar, Gautam, and Joshi (2013) for a more detailed       86	 Both research and extension efforts (using annual
    description of the results of the econometric analysis,           expenditures as proxies) have strong lagged effects to capture
    including a description of the data, variable definitions,        the lagged effects of research investment, research stock is
    and methodology. Briefly, the results are based on panel          approximated as a weighted sum of research expenditures per
    regressions of TFP on explanatory factors, with state and         hectare over the previous six-year period. Similarly, extension
    year fixed effects to control for unobservable factors. The       stock is approximated as a weighted sum of extension
    panel data comprise TFP estimates from 1981 to 2008 for the       expenditures per hectare over the past three years. The
    15 major states for which crop sector TFP estimates can be        weights assume a lagged effect that rises with time and then
    calculated using the CoC database.                                tapers off (following Evenson, Pray, and Rosegrant 1999).




   70
                                                                        Chapter 7: Total Factor Productivity Growth




stock emerges as a consistent and important                        Another important finding—and cause for
driver across the various specifications                           concern from the perspective of environmental,
estimated. An increase in research stock (that                     productivity, and growth sustainability—is the
is, in expenditures) of 1 percent is estimated to                  excessive or imbalanced use of inputs, which
increase TFP growth by about 0.2 percent. In                       is arguably driven by input subsidies. Two
addition, tests to see if the impact of research                   variables to capture these effects (following
stock has declined over time (that is, if the impact               Chand et al. 2010) are (i) the share of area
in the second half of the sample period—post-                      irrigated by groundwater and (ii) the share
1995—is different from that in the first half)                     of nitrogen in the nutrient mix. Expanding
show that is has not. The impact of research                       groundwater use is closely associated with
in the post-1995 period is not found to be                         electricity subsidies and has proved to be a
statistically different than before, implying that                 double-edged sword. It has contributed to the
once other factors (particularly rainfall) are                     rapid expansion of irrigation and had a large
accounted for, the contribution of technology has                  impact on growth, helping to improve food
not diminished over time.87 This result strongly                   security and reducing poverty. Yet it has also
supports the priority placed by the Twelfth Five                   substantially mined groundwater and led to
Year Plan on public expenditures on research.                      deterioration in the productive resource base
                                                                   (with associated problems of water and soil
In contrast, the impact of agricultural extension is               degradation), especially in the original green
considerably lower and less statistically significant.88           revolution areas. Given these opposing effects,
This low impact is particularly striking, given                    it is not surprising to find that groundwater
that traditional crops traditionally have been                     irrigated area has a statistically insignificant
a priority for public extension. The remaining                     impact on TFP. A telling result, however, is that
large yield gaps (discussed in the next chapter),                  rural electrification (captured as the number
even for the main staples (rice and wheat),                        of villages electrified in a state, normalized
suggest that extension’s impact is not low                         by net sown area to remove the scale effects)
because the current technological possibilities                    has a large, highly significant negative impact.
have been fully exploited—on the contrary,                         Normally one would expect electricity to have
much remains to be done. Extension’s low                           a positive impact as the use of electrically
impact also explains the results from the DEA                      powered irrigation equipment expands, food
decomposition showing stagnant or falling                          processing increases, and so on. Given that most
technical efficiency, implying a widening gap                      rural electricity is consumed for agriculture
between available technology and average                           and used to power the electrical pumps that
practice. To enhance productivity quickly,                         have grown rapidly in number in most states,
increasing the effectiveness of extension is the                   the marked negative impact of electricity is
low-hanging fruit and warrants serious policy                      surprising and strongly suggests that electricity
attention.
                                                                   policies may now be restricting rather than
                                                                   contributing to productivity growth through an
87	 An interaction term between a dummy variable for post-1995     adverse environmental (groundwater) impact.
    and research stock is very small in magnitude and highly
    insignificant.
88	 Extension stock is significant at the 93 percent level, just
                                                                   Similarly, lopsided subsidies for fertilizer,
    below the standard 95 percent level of significance.           which substantially lower the price of urea



                                                                                                                   71
Republic of India: Accelerating Agricultural Productivity Growth




relative to other nutrients and promote its                        correlated with other indicators included in
indiscriminate use, have been argued to create                     the analysis, so while they are important, their
nutrient imbalances with negative impacts on                       effects are captured in the other indicators.90
soil health. The empirical results, showing a negative
and statistically significant impact of fertilizer on TFP,
strongly support this assertion.                                   Implications: Compromising
                                                                   Sustainability
To test for the impact of nonagricultural
                                                                   TFP growth since 2003 has been robust,
investments through output (food processing)
                                                                   reversing the worrisome slowdown since about
or factor markets (labor)—in other words, to
                                                                   the mid-1990s. The main driver of past TFP
test for the effects of backward linkages on
                                                                   growth has been technical change. Efficiency
agricultural productivity—real wages paid by
                                                                   has stagnated over the long run and shown a
agro-industry in the state were included as a
                                                                   declining trend in recent years. An important
potential factor driving TFP. No association was
                                                                   point is that TFP for traditional crops has
established between industrial wages and TFP for
                                                                   shown little improvement, implying that the
traditional crops.89
                                                                   overall sectoral TFP is being driven by high-
                                                                   valued horticulture and livestock subsectors.
Additional hypotheses tested proved to be
                                                                   The stark difference in the performance of
inconclusive or, in one case, counterintuitive.
                                                                   traditional and high-value crops is all the
Road density is found to be insignificant, as
                                                                   more striking given that policy and public
is an index for infrastructure (a weighted
                                                                   expenditures continue to be heavily focused on
combination of road density, rail density, and
                                                                   the traditional crops. Within traditional crops,
electricity consumption in agriculture per
                                                                   the major share of output growth has occurred
hectare). This result may reflect the finding
                                                                   because of increased use of inputs. More
(described in Chapter 2) that road infrastructure
                                                                   detailed analysis shows that subsidy-driven
even in laggard districts has improved to a
                                                                   input use now negatively affects TFP, indicating
level at which it may no longer be a binding
                                                                   that ironically, instead of boosting productivity,
constraint. An even more counterintuitive
                                                                   subsidies might now be contributing to lower
result is the negative and significant effect
                                                                   productivity, compromising sustainability and
of rural literacy. This result is consistent
                                                                   future productivity growth.
with the Bosworth and Collins (2008) growth
decomposition, also discussed earlier, which                       90	 Both variables, when included by themselves with only
shows that literacy (basic education level) has a                      state-fixed effects, show positive, substantial, and significant
                                                                       positive effects (with elasticities of 9 percent and 4 percent,
limited contribution to sector growth. Another                         respectively), but these effects vanish when other variables
plausible reason for these results (both for roads                     are included. Additional tests show road density to have
                                                                       significant nonlinear (quadratic) effects, suggesting positive
and literacy) is that these variables are highly                       impacts at low levels of investment that diminish with
                                                                       additional investment. The estimated results show a
89	 Agro-industry wages and industrial capital stock (to reflect       maximum impact is reached at about 500 kilometers per
    past investments) also had no significant impact.                  square kilometer of surface area.




   72
8                       Productivity and Efficiency at
                        the Household (Micro) Level




D
        etailed panel data from the NCAER-                          to separate the change in productivity into two
        REDS household surveys provide a                            mutually exclusive and exhaustive components:
        rare view of changes in agricultural                        Efficiency Change (EC) and Technical
inputs and outputs among farm households                            Change (TC).
between 1999 and 2007 and help to ground the
understanding of absolute levels and sources of                     Absolute levels of efficiency for each year
inefficiency in Indian agricultural production.                     are estimated as well. Two measures of
The analysis in this chapter uses nonparametric                     efficiency, with different policy implications,
as well as parametric approaches to gain                            are calculated—technical and economic
additional perspective on trends in agricultural                    efficiency.92 The former is a pure input-output
productivity at the farm level.                                     based (technical) measure indicating whether
                                                                    farmers are using the available technology
                                                                    efficiently. The latter is a measure of economic
Drivers of Changes in Productivity:                                 performance, indicating whether the inputs
Nonparametric Analysis                                              used and outputs produced are economically
                                                                    optimal.93 The analysis also estimates the cost
The change in productivity at the farm
                                                                    of inefficiency in terms of profits lost owing to
level between 1999 and 2007 is measured
                                                                    different types of inefficiencies.
by Nin-Pratt and Gautam (2013) using Data
Envelopment Analysis (DEA) to derive
                                                                    The analysis is done at the whole farm level.
the Malmquist index.91 The advantage of
                                                                    As such, the observed technical change may
the methodology is that it does not entail
assumptions about economic behavior (profit                         92	 To define the input-based Malmquist index, production
maximization or cost minimization) and does                             technology and production efficiency must be characterized.
                                                                        Griliches (1964) defined technology as “the currently known
not require prices for estimating productivity                          ways of converting resources into outputs.” Farrel (1957)
change. The methodology also makes it possible                          introduced the concept of efficiency, which is defined as the
                                                                        ability to produce “the maximum amount of output that is
                                                                        physically achievable with current technology.”
91	 The Malmquist index measures the TFP change between two         93	 Technical efficiency measures the distance at which each
    data points (such as data for a farmer at two points of time)       farmer is operating from the technical frontier (determined
    by calculating the ratio of the distance of each data point         as the maximum output obtained by any farmer from within
    relative to a common technological frontier. The Malmquist          the sample). Economic efficiency is measured as short-run
    index, pioneered by Caves, Christensen, and Diewert                 profit (revenue minus variable costs) for each household,
    (1982), has been extensively used to measure and analyze            which is then compared to the maximum profit that could be
    productivity since Färe et al. (1994) showed that the index         achieved with the given factor endowments of the household
    could be estimated using DEA.                                       (land, family labor, and assets).




                                                                                                                               73
Republic of India: Accelerating Agricultural Productivity Growth




reflect changes in the output mix or the input                    Average technical efficiency at the
                                                       Table 14:  
mix, perhaps as households shift to crops that                    regional and country level
are more productive or intensify the use of                                                          Country-wide
inputs that are more productive. Change may                             Regional efficiency        (meta) efficiency
also arise from a purely technological advance          Region           1999       2007            1999       2007
that is neutral across outputs and inputs. To           Humid            0.552      0.500           0.463      0.418
identify the source of technical change, it is          Semi-arid        0.523      0.544           0.456      0.469
decomposed into three components: (i) output-           temperate
biased technical change (OBTC); (ii) input-biased       Semi-arid         0.552        0.541        0.411        0.451
technical change (IBTC); and (iii) a magnitude          tropical
or neutral component (MTC). Efficiency is               Arid              0.753        0.650        0.436        0.386
decomposed into two components: (i) pure               Source:  Nin-Pratt and Gautam 2013.
technical efficiency (PEC) and (ii) scale efficiency
(SEC). Scale efficiency is measured relative to
                                                       needs to “catch up” to the meta-frontier.94 The
the “optimal scale,” which is associated with the
                                                       Technology Gap Ratio reflects the difference
farm having the highest output to input ratio
                                                       between the regional and meta-frontiers and
(average product), assuming “constant returns
                                                       indicates the state of technology in each region
to scale” (CRS).
                                                       relative the rest of India at a point in time.
To account for India’s agro-ecological
heterogeneity, productivity changes are                Results
estimated by agro-ecological regions, based on         The technical efficiency levels for each year are
the ICRISAT classification of districts across         given in Table 14, which shows the averages
India into four zones (Table 13). A regional           for the four major agro-ecological zones with
Malmquist index measures productivity change           respect to both the regional and the meta-
with respect to the regional frontier (Mr),            frontiers.95 The average farmer’s level of
and a national Malmquist index measures it             technical efficiency is estimated at 50 percent
with respect to the country-wide frontier or           with respect to the meta-frontier. The arid zone
meta-frontier. The differential change in the          seems more homogenous; most farmers are
two frontiers shows how much each region               closer to the regional frontier (average regional
                                                       efficiency is high at 75 percent). The efficiency
Table 13:  Regional classification                     relative to the meta-frontier is significantly
             Agro-ecological                           lower, indicating that the regional frontier
Region        classification         Characteristics   itself reflects lower productivity compared to
1          Humid                  LGP: >180 days       the other regions (or the meta-frontier). This is
2          Semi-arid              LGP: 75–179 days;
           temperate              temperature <18° C   94	 The catch-up term is a combination (product) of (i) the
3          Semi-arid tropic       LGP: 75–179 days;        movement of the regional frontier (TCr) relative to that of the
                                  temperature >18° C       meta-frontier and (ii) regional efficiency growth relative to
                                                           the meta-frontier.
4          Arid                   LGP: 0–74 days       95	 The results are from the analysis of a panel of 1,971
                                                           households between 1999 and 2007. Observations that were
Source:  ICRISAT 2005.                                     either outliers or had missing data were eliminated. Results
Note:  LGP = length of growing period.                     presented are averages across the sample households.




    74
                                        Chapter 8: Productivity and Efficiency at the Household (Micro) Level




Figure 45: Technology gap ratio                              years. Table 15 presents a dynamic picture of
1.000                                                        productivity change, providing productivity or
0.900                                                        TFP growth rates for each region as well as the
0.800                                                        decompositions discussed previously.
0.700
0.600
0.500                                                        The two main components of TFP are technical
0.400                                                        change (TC) and efficiency change (EC). Average
0.300
0.200                                                        TFP growth for India during 1999–2007 was 1.73
0.100                                                        percent, with an even higher growth rate for
0.000                                                        technical change (2.12 percent). The striking
        Region 1     Region 2     Region 3     Region 4
                                                             result is the negative growth in efficiency, which
Source: Nin-Pratt and Gautam 2013.                           compromised the growth in technical change
                                                             and reduced growth in TFP. These findings are
not surprising for the water-stressed arid zone.             qualitatively similar to the results discussed in
The relative regional performance is more                    Chapter 7, in which technical change was found
accurately reflected in the larger technology                to be the main driver of productivity growth.
gap for Region 4 relative to the meta-frontier               Technical change was fastest in Regions 3 and 4,
(Figure 45).                                                 which have traditionally lagged in performance.
                                                             Technical change has also been output neutral,
The technical efficiency measures in Table 14                with some minor input bias, especially in Region
are static measures (at a point of time) of                  3 (the semi-arid tropics).
relative performance and do not convey how
performance has changed over time. While                     The poor performance in efficiency is the result
the low levels of relative technical efficiency              of pure inefficiency, as scale efficiency shows a
indicate significant potential for improvement,              slight improvement of 0.44 percent annually.
they do not indicate if any improvement                      Regions 1 and 4 appear to have experienced a
in productivity occurred between the two                     significant fall in average efficiency, implying

Table 15: TFP growth decomposition (average growth rate, 1999–2007)
                         Region 1            Region 2           Region 3            Region 4               India
TFP                        0.75                1.17               3.16               -0.36                 1.73
TC                         1.89                1.06               2.87                2.79                 2.12
  MAGTC                    1.30                0.63               1.96                2.40                 1.47
  OBTC                     0.06                0.13               0.07                0.09                 0.08
  IBTC                     0.53                0.30               0.92                0.29                 0.51
EC                        -1.12                0.11               0.29               -3.06                -0.38
  SEC                      1.03               -0.22               0.93               -1.77                 0.44
  PEC                     -2.13                0.34              -0.63               -1.31                -0.82
Source:  Nin-Pratt and Gautam 2013.
Note:  
       Decomposition is exact for multiplicative indices; growth rates do not add up exactly. TC = technical change;
       MAGTC = magnitude of technical change; OBTC = output-biased technical change; IBTC = input-biased technical
       change; EC = efficiency change; SEC = scale efficiency; and PEC = pure technical efficiency.




                                                                                                                   75
Republic of India: Accelerating Agricultural Productivity Growth




that the average farmer is moving farther from             most vibrant region, catching up with the
the frontier. The rising disparity in productivity         meta-frontier at a very rapid rate (1.97 percent).
across households is most significant in the               Region 1 fell modestly behind the country-wide
arid areas (Region 4). Given the strong growth             frontier (-0.14 percent). Some of these dynamics
in technical change, this result suggests that             are explained by looking at the technology
some farmers gained access to irrigation (the              gap ratio, which shows that Region 4 has
share of irrigated area has risen in this region)          historically lagged behind the other regions in
and hence to a new production frontier,                    terms of technology (Figure 45).
while the remaining (rainfed) farmers reflect
lower relative efficiency with respect to this             The high cost of inefficiency in
new frontier. Regions 2 and 3 maintained the               agricultural production
average distance from the frontier, which is a
relatively better performance, considering that            The analysis of productivity growth and its
the frontier was moving outward as well.                   components is based on physical inputs and
                                                           outputs and gives a picture of the technical
Region 3 (the semi-arid tropical zone) shows by            changes taking place at the farm level. What
far the best TFP performance—TFP grew at 3.16              do those technical changes mean to the welfare
percent per year—with the fastest growth in                of the farmer in terms of net returns or profits
technical change among regions and a modest                from farming? Estimates of economic efficiency
improvement in efficiency, mostly explained by             provide an answer to this question and indicate
scale efficiency. This region also had the highest         the extent of losses associated with two types
input-biased technical change. In Region 4, the            of inefficiencies: technical inefficiency and
technological frontier is moving fast, but most            allocative inefficiency.
producers are falling behind. Growth in Regions
1 and 2 involved less technical change.                    Table 16 shows results from the NCAER-REDS
                                                           household panel survey for 1999 and 2007. It is
The dynamics of regional and country-                      important to note that the figures presented in
wide frontiers show that Region 4 was the                  the table are percentages of profits lost relative

Table 16: Proportion of profit lost in relation to short-run potential profit
                        1999                                  2007                      Change (%)
                    Profits      Profits                 Profits      Profits           Profits      Profits
Farm       Total lost due to lost due to     Total     lost due to lost due to Total lost due to lost due to
size       profit technical    allocative    profit     technical   allocative profit technical    allocative
quintile    loss inefficiency inefficiency    loss    inefficiency inefficiency loss inefficiency inefficiency
1st         0.59     0.51         0.08        0.59         0.46        0.13      0.25   -10.49       72.80
2nd         0.71     0.51         0.20        0.72         0.45        0.27      1.30   -11.73       35.47
3rd         0.73     0.47         0.25        0.68         0.46        0.22      -6.24   -3.20       -11.93
4th         0.71     0.39         0.31        0.74         0.42        0.32       4.84    6.99        2.15
5th         0.75     0.31         0.44        0.66         0.34        0.33     -11.60    7.55       -25.36
India       0.73     0.36         0.37        0.68         0.38        0.31      -7.05    3.95       -17.75
Source: Nin-Pratt and Gautam 2013.




  76
                                   Chapter 8: Productivity and Efficiency at the Household (Micro) Level




to the optimal economic profit that is feasible.     average technical efficiency as well as a negative
Results are grouped by quintile of farm size         trend between the two survey years, driven
(measured as total cropped area) to assess how       largely by rising disparities in growth within
inefficiencies change with changes in the size of    the humid and arid agro-ecological zones (the
farm operations.                                     former concentrated in Eastern India and the
                                                     latter in the Western part of the country).
On average, sample households lost a huge            Third, current levels of efficiency are associated
73 percent of potential short-run profits in         with very high economic costs, as indicated
1999; they lost somewhat less (68 percent) in        by 68 percent of potential profits lost due to
2007. Losses owing to technical and allocative       technical and economic inefficiencies (down
inefficiencies are about the same (37 percent        from 73 percent in 1999). Finally, small farmers
and 36 percent, respectively). Between 1999          exhibit higher technical inefficiency than large
and 2007, inefficiency was reduced marginally        farmers, perhaps reflecting patterns in access to
by 7.05 percent, mostly because of improved          extension or technology services, but they also
allocative efficiency (17.75 percent). Technical     show lower allocative inefficiency, reflecting
inefficiency worsened marginally (as indicated       more rational economic decisions.
by the earlier decomposition of TFP).
                                                     Changes in Efficiency over Time:
Small and large farms (the 1st and 5th quintiles)
have very different sources of inefficiency.         Parametric Analysis
Smaller producers are more efficient than            The econometric analysis uses a stochastic
large producers (in other words, their foregone      frontier approach to statistically derive the
profits were smaller) because their allocative       results on technical and allocative efficiency
efficiency was significantly higher (with an         for each household in the same NCAER-REDS
associated profit loss of only 8 percent in          sample. This analysis is done for several reasons.
1999). Larger producers showed high allocative       First, it helps to verify the main findings of
inefficiency but were relatively more technically    the nonparametric analysis by moving beyond
efficient than smallholders. The change in           the whole-farm level and to the crop level.
efficiency between 1999 and 2007 is explained        Second, the analysis can be extended to a longer
by improvements in efficiency among the              timeframe by using the 1982 survey, providing
group of large producers (5th quintile). Smaller     insight into farmers’ performance over 25 years
producers did not improve overall efficiency,        and overcoming the potential limitations
and their allocative inefficiency worsened even      associated with the 1997–2003 period (the
as they improved their technical efficiency.         slowdown in growth caused by vagaries of
                                                     weather, discussed in Chapters 2 and 3).96 A
                                                     third advantage of this approach is that it can
Major findings
                                                     take advantage of the flexibility of statistical
Four major findings emerge from the DEA
(nonparametric) analysis. First, technical           96	 The 1982 survey was more limited and the 2007 was the most
change made a strong contribution to growth,             detailed in terms of data collected (and available). Making the
                                                         best use of available and comparable data, basic results are
reflecting the impact of technological advances.         estimated for all three years, with deeper analysis restricted
Second, the analysis found a low level of                to the more comprehensive 2007 data.




                                                                                                                 77
Republic of India: Accelerating Agricultural Productivity Growth




            echnical efficiency distributions:
Figure 46: T                                                         Changes in farm-level technical
                                                          Figure 47: 
           1982, 1999, 2007                                          efficiency by state
8                                                           2                                               .76
                                                                                                            .74
                                                           1.5
6                                                                                                           .72
                                                             1                                              .70
                                                                                                            .68
4
                                                           0.5                                              .66
                                                                                                            .64
2                                                            0
                                                                                                            .62
                                                               RJ OR MP KT KE UP TN MH HR GJ AP WB HP PJ BH
                                                          -0.5                                              .60
0                                                                                                           .58
    .0      .2          .4           .6         .8    1     -1                                              .56
                 Farm Level Technical Efﬁciency                                                             .54
                                                          -1.5     1982-99   1999-07   1982-07   1982 Level
                  1982          1999          2007

Source: Authors, using NCAER-REDS Household panel data.   Source: 
                                                                  Authors, using NCAER-REDS Household Panel data.


analysis and extend the analysis to identify              the states by increasing level of average baseline
potential determinants of inefficiency. Finally,          (1982) technical efficiency (dotted line, right
it is useful for triangulation and for overcoming         axis). Farmers in Bihar represent the case
methodological issues associated with DEA                 of the “poor but efficient” producer, in that
(sensitivity to outliers and measurement errors).         they operate at low input-output levels, but
                                                          their input use was efficient in 1982.97 Punjabi
Technical efficiency at the household                     farmers are equally efficient—but represent the
                                                          high input-output case.
(whole-farm) level
Parametric estimates of technical efficiency at           Figure 47 also shows annual growth in
the household (whole-farm) level are generally            technical efficiency for the entire period
higher than those estimated by the DEA method             (1982–2007), as well as changes between the
for 1999 and 2007 (Figure 46). The average level          three survey rounds (1982–1999 and 1999–
of technical efficiency, shown by the vertical lines      2007). Most states show an improvement in
corresponding to each survey year in Figure 46,           technical efficiency in the first period (with
increased from 68 percent in 1982 to 78 percent           the sole exception of Himachal Pradesh),
in 1999, indicating that on average households            but a majority show a decline in efficiency
moved significantly closer to the (respective)            in the second period. The decline from 1999
frontier. The changes in the distribution of              to 2007 is pronounced for Andhra Pradesh,
efficiency also show convergence among                    Bihar, Odisha, Punjab, Tamil Nadu, and West
households over the long term. The findings               Bengal. The absence of any obvious agro-
mirror the DEA finding that technical efficiency          ecological or geographical correlation in terms
fell modestly between 1999 and 2007. The decline          of performance suggests that policies and the
in efficiency in 2007 is across the board—the
                                                          97	 Recall that technical efficiency measures the distance of each
entire distribution shifts to lower levels.                   farm from the frontier for the level of inputs being used. A
                                                              farmer using low levels of inputs may be operating at the
                                                              low end of the production possibility frontier but may still
Changes in farm-level technical efficiency vary               be relatively efficient (in terms of the distance to the frontier
significantly across states. Figure 47 presents               corresponding to that level of inputs).



    78
                                                Chapter 8: Productivity and Efficiency at the Household (Micro) Level




enabling environment at the state level may                           1999 and 2007 is the group of “other crops”
play a strong role in determining efficiency                          (all crops other than the cereals, pulses, and
levels. For the full period (1982–2007),                              oilseeds). For all other groups, efficiency
efficiency growth is positive for all states,                         is lower or the same. Notably, efficiency is
although it varies inversely with the base year                       lessening in the most widely grown crops (rice
(1982) levels of efficiency.                                          and wheat) and declining significantly for
                                                                      pulses and oilseeds. Across households there
Technical efficiency at the crop level                                is significant dispersion in efficiency levels for
                                                                      all crops, but particularly pulses, oilseeds, and
The data also allow more in-depth assessment                          “other” crops.
at the crop level. Crops are classified into broad
groups with the exception of wheat and rice,
                                                                      Regional differences in efficiency
the most widely grown crops. An estimate
across all crops/plots is also derived, showing                       The wide dispersion across households may
the average (meta) efficiency across crops. The                       reflect variation in geographical or agro-
average efficiency levels are plotted in Figure 48                    ecological conditions. The distribution of
(left panel) and with their distribution across                       efficiency levels by broad geographical
all households (right panel). As expected, the                        classification (North, South, East, and West)
meta-efficiency is representative of individual                       shows that farmers in northern India are
crop efficiencies, showing on average a lower                         performing much better than those in the South
level of efficiency than the main staples (rice                       (Figure 49). The East and West are in between,
and wheat).                                                           with western India performing relatively better
                                                                      than eastern India. The distribution of efficiency
The only group that shows a marginal                                  levels by the four major agro-ecological zones
improvement in technical efficiency between                           shows the expected patterns, with the arid

Figure 48: Crop-level efficiency estimates: 1999–2007

                                                              6
                             Rice
                       0.9

      Meta             0.8                                    4
   (All Crops)          0.7                     Wheat

                        0.6
                                                              2
                        0.5
                        0.4
    Other                                         Other
                                                              0
    Crops                                         Cereals
                                                                  0           .2          .4        .6            .8      1
                                                                                      Technical Efﬁciency
                                                                              Rice                          Wheat
            Oilseeds                   Pulses                                 Other Cereals                 Pulses
                                                                              Oilseeds                      Other Crops
                       1999         2007                                      Meta (all crops)

Source: Authors, using NCAER-REDS Household Panel data.




                                                                                                                          79
Republic of India: Accelerating Agricultural Productivity Growth




Figure 49: Regional differences in efficiency by geographical and agro-ecological zone
                 8




                                                                                   6
                 6
Kernel Density




                                                                  Kernel Density
                                                                                   4
                 4




                                                                                   2
                 2
                 0




                                                                                   0
                      .2   .4               .6           .8   1                        .2            .4           .6              .8             1
                                Meta-efﬁciency 2007                                                         Meta-efﬁciency 2007

                                    North         East                                       Humid                          Semi-arid Temperate
                                    South         West                                       Semi-arid Tropical             Arid


Source: Authors, using NCAER-REDS Household Panel data.


zone (mostly the extreme western part of the                      These telling results suggest that the meager
country) at the lowest levels and the semi-arid                   resources at the disposal of the primarily small
temperate zone at the highest levels (mostly                      farmers in the sample (the median farming
the northern part of the country). The semi-arid                  household has 2.75 acres of owned land and
tropics (much of western, central, and southern                   3.75 acres of operated land) are not being used
India) perform less well than the humid zone                      efficiently, despite relatively high levels of
(on the eastern and western coasts).                              technical efficiency.

Concerns with economic and allocative                             Another important finding is the difference in
efficiency at the farm and crop level                             efficiency levels at the crop level compared to

The striking finding from the DEA analysis was                               Farm economic and allocative
                                                                  Figure 50: 
the extremely low levels of economic efficiency,                             efficiency
which translated to huge profit losses. Using the
                                                                                       0.9
estimated frontier cost function for the years                                         0.8
1999 and 2007 (no price data are available for                                         0.7
                                                                                       0.6
1982), the economic efficiency estimated from                                          0.5
the parametric approach provides qualitatively                                         0.4
                                                                                       0.3
similar results to the DEA analysis. Figure 50                                         0.2
summarizes the household-level estimates.                                              0.1                                 0.18               0.17
                                                                                         0
Estimated economic efficiency was extremely                                                       1982             1999                2007
low (23 percent) in 1999 and declined further                                 Technical           0.68              0.78               0.77
(to 22 percent) in 2007. The relatively high                                  Allocative                            0.23               0.22
                                                                              Economic                              0.18               0.17
technical efficiency levels imply that allocative
efficiency on average is only 17 percent in 2007.                 Source: Authors, using NCAER-REDS Household Panel data.




                 80
                                              Chapter 8: Productivity and Efficiency at the Household (Micro) Level




            llocative efficiency at the crop and
Figure 51: A                                                       productivity problem has been very much on
           farm levels                                             the technical side—either on research to raise
                                                                   yields or extension to deliver research findings
                                                                   to farmers.
        0.91                                      0.98
                 0.78      0.87     0.87                    0.85
  0.8                                                              Without a doubt, technical issues are, and will
                    0.85     0.87          0.95          0.81
          0.71                                                     remain, very important to increase and sustain
                                                     0.23          growth. Yet the findings suggest that it is also
                                                                   critically important to address issues affecting
                                                     0.22
                                                                   the economic rationality of farmers’ decisions.
  Paddy Wheat Other Pulses Oilseeds Other                          In other words, the potential role of policies,
              Cereals               Crops                          institutions, and the enabling environment
 1999      2007     1999 Farm Level          2007 Farm Level       needs be brought to the front and center of the
                                                                   productivity debate if India’s farmers are to
Source: Authors, using NCAER-REDS Household Panel data.
                                                                   move to more profitable agriculture.
the whole-farm level (Figure 51). Estimates of
allocative efficiency show that at the individual                  Determinants of Productivity and
crop level, households appear to be using
                                                                   Technical Efficiency
resources (inputs) reasonably efficiently, given
the output and input prices they face. At the                      What is driving inefficiency at the farm
whole-farm level, however, economic (and                           level? Analysis of the factors influencing
allocative) efficiency falls sharply, suggesting                   land productivity (defined as value of crop
that that the major source of inefficiency                         production per hectare) and the simultaneous
seems to be farmers’ crop choices. Household                       estimation of technical efficiency in the
decisions to allocate resources such as land to                    stochastic production function frameworks
different crops may be an outcome of the policy                    can identify some of the drivers/correlates of
environment (distorted prices, for example),                       the observed inefficiency among farmers.98 The
the higher risks associated with growing the                       main inferences for policy are summarized
more profitable crops, or household preferences                    here; for details see Gautam, Pradhan, and
(to ensure food security, for example). It                         Nagarajan (2013).99
is important that further research clarify
these issues.                                                      98	 The stochastic production function is estimated using a
                                                                       Cobb-Douglas functional form and includes regional and
These findings have direct implications for                            agro-ecological dummy variables. The specification includes
                                                                       nonlinear rainfall effects to adequately control for too
policy makers and researchers. Much of the                             much or too little rainfall, proportion of area irrigated,
debate on Indian agricultural productivity has                         and standard inputs such as seed, fertilizer, pesticide, fixed
                                                                       farm capital, and labor (both human and bullock labor). All
focused on technical or production issues,                             dependent and independent variables are in value per acre
perhaps reflecting the “tonnage” problem—the                           terms. The efficiency function is estimated simultaneously
                                                                       using a number of household, farm, and village or regional
overriding concern with meeting quantitative                           characteristics.
production targets, especially for food                            99	 The results discussed here are based on crop-level estimation,
                                                                       focusing mainly on the meta-production function at the plot
grains (cereals, but lately also pulses) (Sen                          level. The individual results vary by crop to some degree, but
2011). Accordingly, the focus on solving the                           the main implications and conclusions are consistent.




                                                                                                                               81
Republic of India: Accelerating Agricultural Productivity Growth




Major factors influencing crop                       rice, wheat, and oilseed crops and in the meta-
productivity                                         estimation (across all plots). This relationship
                                                     is not found for other cereals, pulses, or the
At the plot level, yields are inversely correlated   broad category of “other” crops. So for the
with plot size for all crops except minor cereals    major crops, and overall, even though larger
and oilseeds. This finding is consistent with the    plots are less productive, larger farms are
long-held technical finding of an inverse land-      more technically efficient. For that reason, it is
productivity relationship.                           important to distinguish the size of individual
                                                     plots from farm size, which may reflect other
Family labor is less productive (at the margin)      advantages that larger farmers may have.
relative to total labor used, implying overuse
or suboptimal use of family labor compared           The analysis provides robust evidence that
to hired labor. The lower marginal labor             the subdivision of plots is already reducing
productivity of family labor is consistent with      productivity. The share of each fragment
the hypothesis that too many people (family          devoted to a particular crop (except for pulses)
members) remain on the farm because of the           is positively associated with technical efficiency.
sluggish structural transformation.                  This result is statistically significant for all
                                                     crops.
Finally, for several crops and in the aggregate,
manure is less productive than inorganic             Infrastructure, as may be expected, has a
fertilizers. Long-term and other benefits may        strong impact, with a consistent finding that
be associated with organic fertilizers such as       the distance of the farm to a pucca (paved) road
manure, but in the short run there is a trade-off    is negatively associated with efficiency. With
in terms of lower productivity.                      infrastructure accounted for, distance to market
                                                     has no further impact.100
Major factors affecting technical
efficiency                                           Considering the importance of extension
                                                     services to technical efficiency, the analysis
The analysis of demographic variables shows
                                                     provides interesting results. Two variables are
households with older heads to be generally
                                                     used—one is distance to a public extension
less efficient (with the exception of wheat and
                                                     worker, and the other is the number of times
pulse producers, who are more efficient). The
                                                     the farmer has participated in extension
more educated households tend to be more
                                                     activities, such as a demonstration, mela (fair),
efficient. The implication is that a younger,
                                                     or other event. Of the two, public events
better-educated generation will be an important
                                                     like demonstrations have a very strong and
driver of growth in productivity—if they ever
                                                     consistent impact on improving efficiency,
gain access to land to put their energy and skills
                                                     indicating the importance of information
to work.
                                                     and practical demonstrations of technology.
                                                     Distance to extension worker has a mixed
Another important result is the positive
                                                     impact. Proximity to an extension workers is
relationship between farm size (total land
owned by the household) and technical                100	 It is statistically significant only for two crops, but with
efficiency, which is statistically significant for        mixed influence.




  82
                                   Chapter 8: Productivity and Efficiency at the Household (Micro) Level




negatively associated with efficiency for wheat,     SC farmers are less efficient in rice and wheat,
but positively associated with efficiency for        and ST farmers are less efficient in wheat
minor cereals, oilseeds, and “other” crops. The      production. This continued inefficiency for the
impact on wheat may be associated with the           two crops that receive the most public attention
de facto role of public extension in the green       is surprising and calls for a focus on the
revolution areas, where extension has been           inclusiveness agenda for extension services. The
associated with distributing subsidized inputs       lack of significance in “other” crops could mean
and possibly overuse of inputs to the detriment      either insufficient statistical power (due to a
of technical efficiency.                             small number of observations or wide variation
                                                     in performance), or it could mean that the
Villages that are more technologically advanced      “other” (relatively minor) crops are part of the
(using the proportion of area planted to HYVs        traditional portfolio of disadvantaged groups,
as a proxy) are strongly and significantly more      who may be more proficient in growing them.
efficient. The importance of technology in all
crops and overall is amply clear.
                                                     Changes in Agricultural Relations:
Tubewells, which allow much better control of        Are Smaller Farms Still Efficient
irrigation water, contribute to efficiency. Access   and Viable?
to canal water is less consistent: It contributes
                                                     Before concluding the discussion on
to efficiency in rice and minor cereals, and
                                                     productivity and efficiency, it is vital to
across all crops, but it is not significant for      understand how the underlying economic
“other” crops.                                       relationships are changing and affecting
                                                     farmers’ incomes and well-being. The variables
Governance indicators provide additional             of interest are farm size, in particular land
important insights. A strong and consistent          owned by households, and the returns to
result across all crops is the strong impact on      farming. Three levels of farming returns are
efficiency of women’s reservation (assured           used: gross value of output per hectare, gross-
participation) in Gram Sabha meetings. A clear       margins per hectare (revenues less paid out
message is that involvement of women in local        costs, so in essence returns to family labor and
governance is good for productivity growth           land), and profits (revenues less all costs, that
(and hence prosperity). The proportion of local      is paid costs and imputed costs of family labor).
government expenditures on agriculture has a         Once again, this analysis takes advantage of the
positive impact for the meta-function and for        household panel data from the NCAER surveys
rice and “other” crops, but it is not significant    spanning 25 years from 1982, 1999 and 2007.
for “other” crops.                                   The main findings are presented using simple
                                                     graphics of nonparametric trends to highlight
The findings on social inclusion highlight           the key messages.101
some limited impacts but are not significant
for most crops. Female-headed households are
more efficient producers of oilseeds only and        101	 To focus on the main trends and avoid the noise that may be
                                                          associated with extreme values, the sample is trimmed by
are not more or less efficient for other crops. In        deleting the top and bottom 1 percent of the land (farm-size)
terms of social classes, the results suggest that         distribution.




                                                                                                                 83
 Republic of India: Accelerating Agricultural Productivity Growth




             hanging farm size and profitability
 Figure 52: C                                                              farm size and net revenues or profits per
            relationship, 1982–2007                                        acre (Figure 52). The figure clearly illustrates
                                                                           a progressive change in agrarian relations
2000 3000 4000 5000 6000




                                                                           over time, from a strong inverse relationship
                                                                           between profits (net income earned from crop
                                                                           farming) and farm size to an increasingly
                                                                           positive relationship in 1999 and 2007.102

                                                                           Given that the majority of farms in India are
                                                                           less than 1 hectare (roughly 2.5 acres), these
                                                                           findings are significant—politically as well
                                                                           as economically—for growth and poverty
                                                                           reduction. The argument of “small and efficient
                           0       10       20       30      40       50   (in terms of productivity)” still applies from a
                                Farm Proﬁts 1982   Farm Proﬁts 1999        technical perspective of higher yields (gross
                                Farm Proﬁts 2007
                                                                           output or value of output per acre), but the
                                                                           changing economic relationship indicates
 Source: Authors, using NCAER-REDS Household Panel data.                   that “small and efficient” does not necessarily
                                                                           translate to economic well-being in terms of
 There has been a significant change in the                                more income (profits). The reasons behind the
 structure of agriculture over the 25 years                                shift in the profit and farm-size relationship are
 between 1982 and 2007. Figures A.8.1 and A.8.2                            clearly seen in Figures 53 and 54.
 (Annex 8) confirm two main findings. One is
 the inverse relationship of yields (calculated                            Figure 54 shows that while paid out costs
 as gross revenue per acre) and farm size. This                            rise with farm size, the relationship changes
 relationship has held over the years. Using the                           completely with total costs (that is, when family
 “Situation Assessment Survey of Farmers, 2003,”                           labor is included), which fall with farm size. The
 Chand, Prasanna and Singh (2011) also confirm                             trends in the components of production costs
 the inverse relationship between farm size and                            in Figure 53 depict this point more clearly. In
 productivity (using value of crop output per                              effect, the higher revenues (at the small-farm
 hectare as the indicator of productivity). But the                        level, which turn to almost constant returns
 relationship of technical efficiency with farm                            to scale beyond about 5 acres) are neutralized
 size has changed. For 2006 (the last year of the                          by very high family labor costs.103 These trends
 survey), the trend confirms the econometric
                                                                           102	 Rawal and Swaminathan (2012), based on a study of seven
 finding presented earlier that efficiency                                      villages in three states (Andhra Pradesh, Maharashtra, and
 increases with farm size. For the previous years,                              Uttar Pradesh), find a positive relationship between gross as
                                                                                well as net incomes from farming and the scale of operation
 however, the relationship was nonlinear—rising                                 defined as the “value of means of production,” but no clear-
 slightly from very small farm sizes but then                                   cut patterns emerge between farm size and net income per
                                                                                hectare.
 eventually falling.                                                       103	 A similar conclusion is reached by Chand et al. (2011), who
                                                                                consider output value per capita and find it to be positively
 The most significant and dramatic shift has                                    correlated with farm size—an alternative way of looking at
                                                                                returns to family labor using family size as a proxy for
 taken place in the relationship between                                        family labor.




                           84
                                                       Chapter 8: Productivity and Efficiency at the Household (Micro) Level




 Figure 53: Revenue and costs per acre by farm size                           Figure 54: Cost structure by farm size
                                                                              .5
10000 12000




                                                                              .4


                                                                              .3
8000




                                                                              .2
6000




                                                                              .1
4000




                                                                              0
2000




              0        10                20              30              40        0            10       20           30             40
                                  Total Cropped Area                                                 Cropped Area
                  Crop value per acre                  Paid-out Costs                  Fertilizer    Other inputs      Hired Labor
                  Total Cost (incl. Family labor)      Pro ts per acre                 Pesticides    Family Labor

 Source: Authors, using NCAER-REDS Household Panel data.                      Source: Authors, using NCAER-REDS Household Panel data.


 reinforce the econometric finding that family                                Past (and present) debates on agricultural
 labor use appears to be less productive on farm,                             productivity have centered primarily on
 suggesting that labor is overused on the farm.                               technical or production issues, reflecting the
 Small farmers use less fertilizer and pesticides                             policy focus on quantitative cereal output
 than other farmers, but the intensity of use                                 growth. Promoting diversification toward a
 for other inputs is more or less constant across                             more economically profitable crop mix would
 farm sizes.104                                                               achieve significant gains through improved
                                                                              economic efficiency in production. To achieve
                                                                              this goal, the role of policies, institutions, and
 Summary and Implications: High                                               the enabling environment needs to be brought
 Cost of Inefficiency                                                         to the front and center of the discussions on
 At the household level, the finding of very                                  achieving faster productivity growth.
 low economic efficiency implies a very high
 cost to farmers in terms of foregone farm                                    Along these lines, the findings on land
 income. The higher technical inefficiency                                    fragmentation, efficiency of younger and more
 among small farmers suggests potentially high                                educated households, and the association of
 returns to extension services, whereas the                                   profitability with farm size are very important.
 higher allocative inefficiency at the farm level                             They suggest that some small farms may be
 among large farmers indicates economically                                   getting too small to remain efficient or viable,
 suboptimal crop choice.                                                      despite the inverse technical relationship
                                                                              between farm size and yield.

 104	 The share of fertilizers in total as well as paid out costs also
      rises with farm size, indicating that small farmers are using
                                                                              The findings lend new urgency to land market
      less fertilizer in both absolute and relative terms.                    reforms and efforts to move labor off the farm. The



                                                                                                                                85
Republic of India: Accelerating Agricultural Productivity Growth




issues concerning land markets in India are        land in the hands of the more productive
not analyzed in depth in this study. They are      farmers, perhaps by improving access to land
already well analyzed, and the need to reform      for younger and more educated farmers.
tenancy laws as well as legalize or otherwise      Perhaps even more important is the potential
loosen restrictions on land lease markets          interaction with the dynamics of structural
is widely acknowledged (for example, see           transformation. The need to attract agricultural
discussions in GOI 2008a, 2011b, 2012;             workers to productive jobs off of the farm
Haque 2012, 2013).                                 is well recognized, and the findings in this
                                                   report suggest that it is even more urgent than
This long-standing issue of reforming the policy   anticipated. Better-functioning land markets
and regulatory framework governing land and        would permit the inefficient or unviable
rental markets is a high priority to sustain       farmers to seek off-farm work without fear of
productivity and growth in farm incomes. More      losing their primary asset or being bogged down
efficient land markets can help to consolidate     in debilitating litigation.




  86
9                                                                 Technology: Yield Gaps and
                                                                  Prospects for Growth




M
         any factors drive growth in agricultural                                                                                                                                                       Past investments in technology appear to have
         productivity, such as investments                                                                                                                                                              made a significant impact, with technical
         in infrastructure, education, and                                                                                                                                                              change being the primary driver of TFP. Yet the
market access. Among these, the central role                                                                                                                                                            green revolution technology may be reaching
of “technology capital” is well documented for                                                                                                                                                          its limits and beginning to show diminishing
India and globally (Evenson and Fuglie 2010;                                                                                                                                                            returns with progressively slower growth in
Chand, Kumar, and Kumar 2011; Fuglie 2012;                                                                                                                                                              the green revolution stronghold states. The
Fan, Hazell and Thorat 1999; Fan, Gulati and                                                                                                                                                            arguments for this “technology fatigue” are
Thorat 2008). The previous chapters provide a                                                                                                                                                           grounded in the yield trends of India’s major
somewhat mixed picture of technology’s role in                                                                                                                                                          crops, which show decreasing growth rates
Indian agriculture. The estimates of extension                                                                                                                                                          since the 1980s (wheat) and early 1990s (rice)
stock and research stock suggest that slow                                                                                                                                                              (Figure 55). Growth rates for both crops have
growth in productivity may be related more to                                                                                                                                                           picked up in recent years, but it may be too
deficiencies in extension than in research.                                                                                                                                                             soon to draw robust conclusions, as weather
                                                                                                                                                                                                        clouds a clear assessment.
           Growth in India’s rice and wheat yields
Figure 55: 
           has slowed (10-year trend rates)                                                                                                                                                             What do these long-term trends mean? They
                                                                                                                                                                                                        may reflect insufficient generation of new
4.5
                                                                                                                                                                                                        technology. They may reflect farmers’ failure
4.0
                                                                                                                                                                                                        to use the newest technology available.
3.5
3.0
                                                                                                                                                                                                        Alternatively, farmers may be using the best
2.5                                                                                                                                                                                                     technology but failing to realize the yields that
2.0                                                                                                                                                                                                     the technology is capable of delivering. The
1.5                                                                                                                                                                                                     most alarming prospect is that the technology
1.0                                                                                                                                                                                                     has reached its genetic limits—not only
0.5                                                                                                                                                                                                     in India, but globally. In fact, global yields
0.0                                                                                                                                                                                                     appear to have plateaued since the mid-1990s
      1980/1981
                  1982/1983
                              1984/1985
                                          1986/1987
                                                      1988/1989
                                                                  1990/1991
                                                                              1992/1993
                                                                                          1994/1995
                                                                                                      1996/1997
                                                                                                                  1998/1999
                                                                                                                              2000/2001
                                                                                                                                          2002/2003
                                                                                                                                                      2004/2005
                                                                                                                                                                  2006/2007
                                                                                                                                                                              2008/2009
                                                                                                                                                                                          2010/2011**




                                                                                                                                                                                                        for wheat and late 1990s for rice, as seen in
                                                                                                                                                                                                        Figure 56, which plots the highest rice and
                                                                                                                                                                                                        wheat yields attained in each year since 1960
                                                                              Wheat                                      Rice
                                                                                                                                                                                                        by any of the world’s top 40 producers of rice
Source: Authors, using data from DES, MOA.                                                                                                                                                              and wheat.



                                                                                                                                                                                                                                                     87
Republic of India: Accelerating Agricultural Productivity Growth




            lobal rice and wheat yields have plateaued
Figure 56: G

                12


                10


                 8


                 6


                 4


                 2


                 0
                     1960/1961
                                 1962/1963
                                             1964/1965
                                                         1966/1967
                                                                     1968/1969
                                                                                 1970/1971
                                                                                             1972/1973
                                                                                                         1974/1975
                                                                                                                     1976/1977
                                                                                                                                 1978/1979
                                                                                                                                             1980/1981
                                                                                                                                                         1982/1983
                                                                                                                                                                     1984/1985
                                                                                                                                                                                 1986/1987
                                                                                                                                                                                             1988/1989
                                                                                                                                                                                                         1990/1991
                                                                                                                                                                                                                     1992/1993
                                                                                                                                                                                                                                 1994/1995
                                                                                                                                                                                                                                             1996/1997
                                                                                                                                                                                                                                                         1998/1999
                                                                                                                                                                                                                                                                     2000/2001
                                                                                                                                                                                                                                                                                 2002/2003
                                                                                                                                                                                                                                                                                             2004/2005
                                                                                                                                                                                                                                                                                                         2006/2007
                                                                                                                                                                                                                                                                                                                     2008/2009
                                                                                                                                                                                                                                                                                                                                 2010/2011
                                                                                 Rice Max Yields (Top 40)                                                                                                                Wheat Max Yields (Top 40)
                                                                                 Trend 2000/01-2011/12                                                                                                                   Trend 1995/96-2011/12


Source: Authors, using FAOSTAT.


The food security implications of slow global                                                                                                                                     First, it looks at the “research problem” (the
growth in cereal yields have raised concern for                                                                                                                                   generation of new technology) by assessing
some time (World Bank 2007b, Cassman 2011;                                                                                                                                        how the production frontier has moved over
Ray et al. 2013). For a country the size of India,                                                                                                                                time and examining prospects for progress.
food security is a high priority, especially given                                                                                                                                Second, it focuses on the “extension problem”
the projected volatility in global markets and                                                                                                                                    through yield gap analysis, in which the
the sheer influence of India on those markets. If                                                                                                                                 differences between yields and agronomic
past trends continue, demand-supply projection                                                                                                                                    potential are estimated to clarify prospects for
models show that India’s food security situation                                                                                                                                  improving yields with the technology that is
probably will not be dire (Joshi and Kumar                                                                                                                                        available now.
2011; GOI 2012; Ganesh-Kumar et al. 2012). If
productivity remains sluggish, however, India
might have to rely on imports (Parikh et al.                                                                                                                                      Current Yield Gaps and Progress in
2011), with a significant impact on global                                                                                                                                        Closing Gaps over Time
markets. The slowing growth in yields is thus
worrisome, regardless of whether the objective                                                                                                                                    Yield gaps are variously defined based on the
is to maintain yield trends for food security or                                                                                                                                  benchmark selected for “potential” yields
to improve productivity for growth.                                                                                                                                               (Fischer, Byerlee and Edmeades 2009; Lobell,
                                                                                                                                                                                  Cassman and Fields 2009). While it is tempting
To assess the prospects for yield growth, this                                                                                                                                    to use yields achieved in other countries or
chapter looks at two aspects of the issue.                                                                                                                                        certain regions within India as the benchmark,



  88
                                                          Chapter 9: Technology: Yield Gaps and Prospects for Growth




such comparisons may not be valid.105 Another                         Figure 57: Yield gaps, all India
approach (adopted here) is to use a bio-physical                                        9
                                                                                              8.2
crop model to simulate the potential for                                                8
popular and established rice and wheat cultivars                                        7
                                                                                                    6.2                        6.6




                                                                      Yields, Tons/ha
in different parts of India.                                                            6                                            5.5
                                                                                        5
                                                                                        4
Two benchmark levels are generated for                                                                    3.4
                                                                                                                                           2.9
                                                                                        3
each state: One is a biological potential yield
                                                                                        2
(PY) based only on varietal characteristics
                                                                                        1
without any constraints in the growing
                                                                                        0
environment, and the other is attainable yield                                              Paddy (All India)              Wheat (All India)
(AY), which introduces water management                                                     Simulated PY        Simulated AY     Actual
as a constraint, using the current level of                           Source: Authors and DES, MOA.
irrigation development in each state to
simulate attainable yield.106 A final benchmark                       Figure 57 shows the all-India weighted averages
is “realizable” yield (RY), which is the research                     for RY, AY, and FY, giving an aggregate picture
station or experimental yield. Experimental                           of yield gaps for rice and wheat. Given India’s
yields (RY) should be close to the AY, with the                       heterogeneity, yields and yield gaps vary by
difference reflecting soil problems, pests, other                     location, making it difficult to arrive at an
management problems, and localized variation                          aggregate RY estimate (RY data are classified only
in weather, as some biotic and abiotic stresses                       by broad agricultural domains). The importance
cannot be controlled even on an experiment                            of this limitation is clearly shown for rice and
station. All of these yields can then be compared                     wheat in Annex 9, Figures A.9.1–A.9.2. Even
to observed actual or farm yields (FY) to                             theoretical potential yields vary considerably by
establish the size of the yield gaps.                                 state, with greater differences for wheat than
105	 Agro-ecological conditions (temperature, sunlight, water
                                                                      for rice. The variation in AY is larger, indicating
     stress, and so on), soil type, and various other biotic          that even within India, a simple comparison of
     and abiotic factors make such comparisons difficult or
     meaningless. For example, it is commonplace to compare
                                                                      rice yields observed in, for example, Punjab with
     rice yields in India (typically Punjab) with China (typically    Madhya Pradesh or Jharkhand may not be useful.
     Jilin or Jiangsu, provinces with highest yields), despite
     marked differences in soil conditions (such as soil carbon,
     micronutrients), temperature, and growing duration (length       Although RY would be the ideal comparator,
     of growing period or exposure to sunlight), all of which make    AY is used as a proxy, as they are expected to
     a significant difference to yields. Farming systems differ as
     well. Rice is typically monocropped in China, and the length     be close. As shown in Figure 58, RY and AY are
     of the growing period is significantly longer than in India’s    very similar for Punjab (where RY is more easily
     Punjab, where farmers grow rice and wheat in rotation. Their
     priority is to shorten the growing period for rice to make way   compiled based on available data). Assuming
     for wheat planting, even though rice yields are compromised      this relationship holds for other states,
     as a result.
106	 The biological potential reflects purely exogenous climate       Figure 59 shows the current yield gap107 for the
     factors and varietal characteristics, but no other biotic
     or abiotic stresses (considered manageable), and hence           107	 The yield gap is defined as the difference between FY and AY
     represents a theoretical physical optimum. Attainable                 as a percentage of AY, where FY is a five-year average actual
     yield incorporates local soil factors and water control (area         or observed farmer yield at the state level, and AY represents
     irrigated) to identify what is attainable given the current           the current simulated potential, with the crop model
     level of water management but assuming that all other                 parameterized using a recently released, widely used variety
     factors can be (at least theoretically) controlled.                   for each crop.




                                                                                                                                                 89
Republic of India: Accelerating Agricultural Productivity Growth




Figure 58: Yield gaps, Punjab                                                                       states show a slowing in yield growth, with
                  12                                                                                the exception of Gujarat, Karnataka, and
                           10.2   9.72 9.7
                                                                                                    Maharashtra, all of which have substantially
                  10
                                                                8.4
                                                                                                    narrowed their yield gaps.
Yields, Tons/Ha




                  8                                                   7.4
                                                                            6.8
                  6
                                             6.2                                                    As a point of reference, it is instructive to note
                                                                                  4.5               that globally no country has been able to reduce
                  4
                                                                                                    its yield gap below 20 percent, suggesting that
                  2                                                                                 30 percent may be a realistic target (Lobell,
                                                                                                    Cassman and Fields 2009). The reasons behind
                  0
                              Paddy (Punjab)                     Wheat (Punjab)                     this phenomenon are not fully understood but
                        Sim. Potential    Sim. Attainable        Var. Rel. Data         Actual      probably result from management and variation
Source: Authors and DES,MOA.                                                                        in field conditions. Yield gaps vary considerably
                                                                                                    across states, but some approach 30 percent of
major producing states. It also shows the rate at                                                   their potential with current technology and so
which the yield gap has reduced over the long                                                       have little scope for further improvement. West
run (1970-2010) and in recent years (1995–2010).                                                    Bengal—not a traditional wheat-growing area—
                                                                                                    is an outlier that has exceeded its expected
Most states have made more rapid progress                                                           performance. Maharashtra and Gujarat seem to
in closing the yield gap for rice in the last 15                                                    have exhausted their potential with the current
years relative to the earlier period (reflected in                                                  wheat technology, whereas Punjab and Haryana
the lower long-run growth rate). The notable                                                        still have some limited room to increase yields.
exceptions are Bihar, Madhya Pradesh, Tamil                                                         West Bengal and Punjab are also close to their
Nadu, and Uttar Pradesh, which have regressed                                                       potential for paddy, but most other states have
in the more recent period. In wheat most                                                            significant potential yield advances to exploit.


Figure 59: Current yield gaps and progress in reducing yield gap
    4.0                                         Wheat                                    70.0 2.5                                Paddy             80.0

                                                                                                 2.0                                               70.0
    3.0                                                                                  60.0
                                                                                                 1.5
                                                                                                                                                   60.0
                                                                                         50.0
    2.0                                                                                          1.0
                                                                                                                                                   50.0
                                                                                         40.0 0.5
    1.0                                                                                                                                            40.0
                                                                                         30.0 0.0
                                                                                                                                                   30.0
    0.0                                                                                          -0.5
                                                                                         20.0
                                                                                                                                                   20.0
                                                                                                 -1.0
 -1.0                                                                                    10.0
                                                                                                 -1.5 WB AP PB TN UP KE KT GJ OR HR MH CH BH MP    10.0
                       WB MH HR      PB    GJ      RJ     UP   KT MP HP           BH
 -2.0                                                                                    0.0     -2.0                                              0.0

                                                        Progress 1970-10     Progress 1995-10           Yield Gap - right axis

Source: Authors, derived using DES, MOA, data.




                  90
                                                             Chapter 9: Technology: Yield Gaps and Prospects for Growth




In interpreting these findings, two important                           early 1990s), yields in rice have not improved
caveats need to be kept in mind. One is that AY,                        significantly in recent years. Longer trends from
used here as the benchmark for the yield gap,                           1965 onward (depicted in Annex 9, Figures
assumes no biotic or abiotic stress is present,                         A.9.3–A.9.6) reveal that growth in RY has slowed
which may not hold in practice—see the                                  in recent years. After the initial sharp growth
discussion on Bihar and Odisha in Chapter 14,                           in the late 1960s with the introduction of HYVs,
for example. Second, these are physical or                              since 1980 rainfed rice yields have changed
output maximum potentials, not economically                             little. For irrigated rice, RY grew fastest in the
optimal potentials. So depending on the local                           1980s and early 1990s and then stagnated from
policy and institutional environment, reducing                          1995. Trends in realizable wheat yields show a
the physical yield gaps may not be economically                         more optimistic picture. Growth increased in
viable (as possibly reflected in the actual or                          the latest period (1995–2010) after steady albeit
observed farm yields).                                                  relatively slower growth from 1965 to 1995.
                                                                        For rainfed wheat, RY increased most sharply
                                                                        in the 1980s; it has slowed since then but
Technological Progress: Evolution                                       continues to grow. Almost 90 percent of India’s
of Realizable Yields                                                    current wheat area is irrigated, however, and
                                                                        after modest but steady growth from the 1960s
Using data on India’s varietal releases,108 it is
                                                                        through the 1980s, potential irrigated wheat
possible to chart the technological progress
                                                                        yields have been on the upswing since 1995.
made for these crops from the 1960s. The
database includes varieties released for 6
                                                                        Figures 60 and 61 also show what is happening
agro-ecological zones for wheat and 12 agro-
                                                                        with yield gaps in rice and wheat at the
ecologies/production systems for rice. It
                                                                        aggregate level. Wheat has seen a recent (1995–
provides information on varieties released
                                                                        2010) widening of the yield gap, whereas the
in every year, including the experiment
                                                                        rice gap has been narrowing. Farm Yields (FY)
station yields that are the basis for varietal
                                                                        are rising each year by 21 kilograms per hectare
release decisions. Several varieties are usually
                                                                        (0.74 percent) for wheat and 26 kilograms per
recommended for multiple domains, and
                                                                        hectare (1.2 percent) for rice. The widening
multiple varieties are released in most years. To
                                                                        yield gap for wheat arises from the much faster
simplify the depiction of the long-term trends,
                                                                        rise in RY (36 kilograms per hectare per year for
Figure 60 and Figure 61 show the maximum
                                                                        rainfed wheat and 54 for irrigated wheat), while
experimental rice and wheat yields across all
                                                                        for rice RY has stagnated since 1995, rising at
varieties released in a given year across all
                                                                        only 6 kilograms per hectare per year.110
domains for irrigated and rainfed conditions.109
The figures also include current farm yields as
                                                                        How does India compare to other countries with
an all-India average.
                                                                        respect to yield gaps for rice and wheat, and
                                                                        progress in closing those gaps? A study of major
The figures show that while wheat potential
                                                                        grain-growing areas around the world (“global
continues to rise (after stagnating in the

108	 Provided by DWR and DRR and facilitated by ICAR.                   110	 Part of the observed growth in FY for each crop is explained
109	 For some years, recorded yields are very low, because of                by irrigation expansion. Between 1995 and 2008 (years for
     special traits in the varieties released. These observations are        which data are available), the share of irrigated area grew at
     dropped to keep the trends from getting too distorted.                  0.5 percent for wheat and 1.04 percent for rice.



                                                                                                                                      91
Republic of India: Accelerating Agricultural Productivity Growth




            xperimental rice yields for released
Figure 60: E                                                                  Experimental wheat yields for released
                                                                   Figure 61: 
           varieties, 1990–2010                                               varieties, 1990–2010
10.00                                                              8.00
 9.00                                                              7.00
 8.00
                                                                   6.00
 7.00
 6.00                                                              5.00
 5.00                                                              4.00
 4.00                                                              3.00
 3.00                                                              2.00
 2.00
                                                                   1.00
 1.00
 0.00                                                              0.00
     1990        1995          2000          2005           2010      1990            1995          2000             2005          2010

        Actual Yields (Combined)   Irrigated PY     Reinfed PY               Actual Yields (Combined)      Irrigated PY     Reinfed PY

Source: Authors, using DRR database.                               Source: Authors, using DWR database.


breadbaskets”) analyzes progress in farm yields,                   the average for other sites, but with almost no
potential yields, and yield gaps for rice and                      growth in RY, India is lagging.
wheat (Fischer, Byerlee, and Edmeades 2013).
The study sites include Punjab for irrigated
wheat and rice and Madhya Pradesh for rainfed                      Metrics of Research Achievement
rice. The findings are summarized in Annex 9,                      Other than Yield
Tables A.9.1–A.9.2. The yield gap in Punjab is
                                                                   Yield is an important metric of research
similar to the average across the other sites,
                                                                   achievements, but it is not the only one.
with scope for doubling current yield levels
                                                                   Research also has to focus on developing
relative to Punjab’s potential yield. Progress
                                                                   varieties that may not necessarily advance
(rate of change) in both FY and PY is also
comparable to other sites. Notably, the progress                   the yield frontier but can cope better with
in wheat RY, estimated from the more complete                      particular stresses in specific environments.
data above, appears to be better than for most                     Developing such varieties is an important part
other countries (sites) in the study.                              of the work undertaken in India’s research
                                                                   system, given India’s vast agro-ecological
The yield gap for irrigated rice in Punjab,                        diversity (Annex 9, Figures A.9.7 and A.9.8 show
however, is larger than the global average.                        rice and wheat varieties released by decade for
Current yields in Punjab need to rise by 75                        specific production domains). Often researchers
percent to close the gap, whereas on average                       develop varieties that are adapted to emerging
global yields need to double. Although Madhya                      conditions (evolving pests and diseases, for
Pradesh performs as well as the only other                         example) or to other local priorities or agro-
rainfed rice site, the scope for improvement                       ecologies, such as resilience to drought, saline
is remains substantial—current rainfed rice                        soils, waterlogged soils, and so on.
yields in Madhya Pradesh need to increase by
150 percent to close the gap. Progress with                        This full set of research priorities and
yields at the farm level in India is the same as                   achievements is not captured through yield



   92
                                           Chapter 9: Technology: Yield Gaps and Prospects for Growth




increases alone. For example, Dar et al. (2012)                 Simulated impact of temperatures on
                                                     Figure 62: 
report that a flood-tolerant rice variety had a                 grain yields
significant positive impact (45 percent higher                          7
yields) under prolonged flooding (6–14 days)                            6
compared to a popular variety in a randomized
                                                                        5
trial in 128 villages in Odisha. Research on




                                                    Grain yield, t/ha
such varieties is a high priority, considering                          4

that 25 percent of India’s rice area is affected                        3
by prolonged flooding and that Odisha’s low-                            2
lying, flood-prone areas are heavily populated
                                                                        1
by disadvantaged groups. In Bihar, another
randomized evaluation of traits in rice varieties                       0
                                                                            0      1          2           3          4            5
(Ward et al. 2013) finds that farmers value                                              Increase in Temperature, °C
drought-tolerant cultivars and are willing to pay                               550 ppm CO2     Potential yield   Current yield
more for them. The high risk of abiotic stresses
                                                     Source: Aggarwal and Rani 2009.
(such as drought) in Bihar warrants additional
research on varieties that can surmount these
                                                     same model used to estimate yield potential
local problems. At the same time, farmers’
                                                     previously), which clearly demonstrates how
willingness to pay for the traits they demand
                                                     rising temperatures might affect grain yields in
in a variety signals that India should encourage
                                                     India.
greater investment by the private sector in
research on such cultivars.
                                                     Technology Capital: Investing for
Another major historical priority of research        the Future
in India has been to reduce the length of
the growing period. As noted, this trait is          The evidence on the role of innovation in
particularly important in the Indo-Gangetic          India’s structural transformation—specifically
plains, where farmers growing rice and wheat         the need to accelerate growth in TFP—
in rotation place a premium on rice that             heightens the importance of investing in
matures early and wheat that can be planted          technology capital. Developing countries now
late. Short-duration varieties are also an           account for almost half of all global spending
important mechanism for managing drought,            on agricultural R&D (Beintema et al. 2012),
because they enable farmers to adjust their          recognizing the need for science and innovation
planting dates and reduce their exposure to          to drive their development. China, Brazil, and
risk. An even more important potential benefit       India are the major powerhouses, together
from varieties that mature early is that they can    contributing about a one-quarter of the global
help farmers adapt to the rising temperatures        R&D investment (Beintema et al. 2012).
that accompany climate change. Higher
temperatures are expected to reduce yields           The recent surge in investment comes in
significantly (Aggarwal and Rani 2009; Jacoby        response to many emerging trends. They
2013). Figure 62 shows the results from recent       include the slowdown in productivity growth
research using a crop simulation model (the          globally and in individual countries—an



                                                                                                                          93
Republic of India: Accelerating Agricultural Productivity Growth




outcome of faltering R&D investments and             in Indian agriculture. Going forward, these
complacency after the success of the green           institutions must internalize the complexities
revolution in most countries, as well as shifting    of the new and emerging food and agricultural
R&D priorities in the developed countries            systems, respond to local demands from
where most technological advances once               smallholders as well as the broad imperatives
originated. Investment is also increasing to cope    of food and nutritional security, feed into the
with greater volatility in global food markets,      rapidly changing and increasingly knowledge-
the prospective impact of climate change             intensive agriculture practiced today, and
on agriculture and food production, and the          provide the high-quality human resources
imperative to ensure future food and nutrition       essential for any technology and innovation
security in the face of severe natural resource      system to succeed.
(land and water) constraints.
                                                     Agricultural research
But success depends on more than the
allocation of public funds, as the comparative       India’s large, well-established agricultural
performance of Brazil, China, and India              research system (Box 2) operates largely in the
demonstrates. Brazil spends significantly less       public domain at both the central and state
than the others but is still widely recognized       levels. Public research still accounts for the
as the most innovative and successful research       bulk of current investment in agricultural R&D.
system among developing countries (Beintema          Rapidly emerging private research operations
et al. 2012). China’s experience shows the           now account for almost one-third of total
importance of focusing on the quality and            R&D investment but need to be nurtured
not just the quantity of innovation effort           through an appropriate enabling environment
(Jin et al. 2010). Successful agricultural           (Pray and Nagarajan 2012). The high returns
innovation systems are built upon strong,            to investments in agricultural R&D are well
interdependent efforts in research, education,       documented worldwide (Alston et al. 2000) and
and extension. All three countries have a long       in India (Rosegrant and Evenson 1992, 1995).
and notable history of achievements in all three     Agricultural research gives higher returns on
components of the innovation system.                 investment than other public expenditures,
                                                     even rural roads and education (Annex 11,
The continuous challenge facing research,            Figure A.11.3) (Fan et al. 2012). The evidence
education, and extension is to remain relevant,      on productivity growth presented in previous
effective, and efficient. Brazil and China have      chapters underscores this point at the national,
undertaken periodic reforms to reorient and          subnational, and farm levels.
redesign their agricultural innovation systems
to meet growing and emerging national                Recognizing the long-term repercussions of
challenges (Huang, Hu and Rozelle 2004;              underinvestment in agricultural R&D in the
Contini et al. 2010). By comparison, research,       1990s, since 2004/05 the government has rapidly
education, and extension have lagged in India        increased public expenditures on agricultural
in recent years (Lele et al. 2013). The impressive   research (including education) at about 15
institutions established at the time of the green    percent per year (Singh 2012). Despite the
revolution have become less effective and            increase in expenditure, agricultural research
relevant in stimulating transformative change        intensity—the ratio of agricultural research


  94
                                                          Chapter 9: Technology: Yield Gaps and Prospects for Growth




Box 2: Agricultural research organization, investment, and capacity in India

The Indian Council of Agricultural Research (ICAR), an autonomous organization under the Department of
Agricultural Research and Education, Ministry of Agriculture, is the apex body for coordinating, guiding,
and managing research and education in agriculture (including horticulture, fisheries, and animal sciences)
throughout India. With 99 ICAR institutes and 60 agricultural universities spread across the country, this
national agricultural research system is one of the world’s largest.
Public spending on agricultural research and education averaged 0.7 percent of agricultural GDP over the
Eleventh Plan period (2007/08–2011/12), compared to an end-of-plan target of 1 percent of agricultural
GDP. Public spending in India is relatively low compared to other developed (2.35 percent) and developing
countries (Brazil spends 1.04 percent).a
The bulk of research capacity, in full-time staff equivalents, is in the state agricultural universities,
accounting for about 56 percent of staff, with the rest in ICAR institutes.b Research resources continue to be
concentrated in the western and southern states (60 percent), with the hill and eastern states receiving less
attention. Research focuses more on crop husbandry (85 percent) than animal husbandry (7 percent) or soil
and water conservation, dairy development, and fisheries (less than 5 percent each).
Private investment in R&D has grown rapidly in the past couple of decades, and in 2008/09 was estimated
to account for 26–30 percent of all spending (public and private) on R&D, compared to about 17 percent
in 1994/95.c The most dynamic sectors for private innovation over the last decade have been the seed
industry, the pesticide industry, and the farm machinery industry. The major remaining constraints noted
by the private sector include regulations and price controls on new technology, enforcement of intellectual
property rights, and limited availability of scientists.
Source: Authors
a
  GOI 2011c. b Beintema et al. 2013. c Pray and Nagarajan 2012.


expenditure to agricultural GDP—remains                              Despite increased spending and rapid growth
very low (under 0.6) (GOI 2012), well below                          in the number of universities and research
India’s comparators and developed countries,                         institutions, the capacity for research in terms
where agriculture’s contribution to GDP and                          of staff strength is falling (Benteima et al. 2012).
employment is much less than in India.                               State agricultural universities are a critical
                                                                     component of the research system, but they
Concerns have grown about the relevance,                             suffer from a number of problems (discussed
efficiency, and effectiveness of the national                        later in detail), including rapid drop in the
agricultural research system, especially its                         number of occupied faculty positions in several
inability to adapt to changing needs and                             state universities and a decline in staff quality
challenges (Evenson and Jha 1973, Jha and                            arising from insularity, ageing, and a general
Kumar 2006; Singh 2011; Lele et al. 2013).111                        decline in skills (Singh 2011).

111	 Key issues include the system’s traditional organizational
     and institutional structure, which is unable to meet the
                                                                     Extensive analysis and two high-powered
     challenges and demands of globalization, privatization, and     committees headed by eminent scientists—the
     liberalization; its technology generation paradigm, which       Swaminathan Committee (2005) and Mashelkar
     is top-down, supply-driven, public-sector-oriented, and
     linear; high transaction costs; deteriorating human capacity;   Committee (2005)—have promoted reforms,
     duplication; bureaucratic rigidities; inadequate incentives     but the general consensus is that the research
     to respond to emerging demands; and poor monitoring and
     evaluation.                                                     system has seen no real reform in how it



                                                                                                                    95
Republic of India: Accelerating Agricultural Productivity Growth




functions (Ferroni 2013; Ramasamy 2012;                  once-impressive university system (Lele and
Lele et al. 2013). A review of contemporary              Goldsmith 1989), patterned after the United
institutional and implementation arrangements            States Land Grant Universities, is in disarray
for agricultural research highlights India’s             and declining in quality. The system faces
need for a strategy to improve the efficiency of         multiple crises of governance, resources,
resource use, increase dissemination of known            effectiveness, and ethics—even as more
technologies, enhance the quality of human               universities are being created, without attention
resources, and commercialize technologies                to quality (Nene and Tamboli 2011). These
(Babu et al. 2012).                                      issues are familiar to the key policy-making
                                                         institutions and decision makers (the Planning
                                                         Commission, ICAR, and Vice-Chancellors and
Agricultural education
                                                         Deans of Universities). Under the auspices of
The state agricultural universities not only             the National Academy of Agricultural Sciences
have a mandate to build the much-needed                  and ICAR, the XI Agriculture Science Congress
human capacity for the research system but               focused on agricultural education, drawing
undertake their own research (particularly               upon best practices across India and from
adaptive research) and extension. India’s                around the world to chart a way forward for

Box 3: Resolutions from the Bhubaneswar Declaration on higher agricultural education in India

The entire community of research, education, and extension professionals, managers, and policy makers
at the XI Agricultural Science Congress (Bhubaneswar, February 2013) unanimously endorsed the following
resolutions from the Bhubaneswar Declaration:
     Embrace agricultural education and AREE4D [agricultural research, education, and extension for
       development] as an integral component of the national agricultural policy to ensure adequate,
       consistent and predictable investments in agriculture, especially education, research, and extension
       in creating a world-class agricultural university system attuned to face challenges and opportunities
       over short, medium and long term.
     Ensure and institutionalize transparent governance, autonomy, meritocracy, dynamic assessment of
       human resource requirement, judicious allocation of resources, effective implementation, monitoring,
       evaluation, accountability and responsibility based system, and to minimize splitting and inbreeding.
     Pay focused attention to the standards, norms, and accreditation in quality agricultural education,
       create centres of excellence and institutes for agricultural education, science, knowledge, research,
       technology, and innovation in an interdisciplinary and multifaculty mode.
     Identify national- and state- level public and private sector leaders with differentiated but reiterative
       responsibilities to work on the design and implementation of reforms and to develop a strong inter-
       ministerial and inter-departmental cooperation mechanism.
     Revamp teaching/learning processes and methodologies to attract best of talents and blooming young
       minds for nurturing them leading to a nation-wide programme on “Youth for Leadership in Farming.”
     Support India’s proposed development of an active and continuous long-term relationship-
       based international cooperation, rejuvenate and dynamically strengthen initial very successful
       collaboration between Indian State Agricultural Universities and United States Land Grant
       Universities, and launch need-based South-South and South-North collaborations such as the Brazilian
       LABEX programme of scientific exchange.
Source: NAAS and ICAR 2013.




  96
                                                          Chapter 9: Technology: Yield Gaps and Prospects for Growth




India. The Bhubaneswar Declaration (NAAS and                          problems in Phase II largely resulted from the
ICAR 2013) outlines a road map for improving                          program’s rapid expansion and inability to
India’s higher education system (Box 3). The                          attract human resources with the appropriate
road map calls for fundamental changes, which                         skills, resulting in a lack of skilled, dedicated
will require political will and commitment to                         personnel, weak research-extension links,
implement and overcome the resistance to                              limited outreach to farmers, limited operational
change.                                                               flexibility (compared to the pilot phase), poor
                                                                      organization, and disappointing outcomes
Agricultural extension                                                (Raabe 2008; Sulaiman and Hall 2008). Despite
                                                                      the revision of ATMA’s guidelines in 2010, a
The primary organizational reform of India’s                          vacuum remains in the delivery of technical
agricultural extension services was the                               advisory services (Ferroni and Zhou 2012). To
gradual implementation of the Agricultural                            assess ATMA’s progress since 2010, Babu et al.
Technology Management Agency (ATMA) model                             (2012) undertook a detailed field assessment
over the past 15 years. ATMA was designed                             based on case studies of seven districts in four
as a decentralized, demand-driven approach                            states (Bihar, Himachal Pradesh, Maharashtra,
to extension that could respond to local                              and Tamil Nadu).114 The study highlighted
demands, priorities, and constraints. Beginning                       positive achievements resulting from ATMA’s
with a pilot (Phase I, 1998–2004), ATMA was                           demand-driven, multi-agency approach
implemented in 28 districts of 7 states and later                     but noted persistent issues with reaching
scaled up (Phase II, 2005–10) to 262 districts and                    communities, achieving organizational
then (2007) to all 591 districts.112 The expansion                    autonomy, and improving staff quality (Box 4).
raised design and implementation concerns,
which were eventually addressed in the revised
guidelines in 2010 ( Annex 9, Box A.9.3) and                          Conclusions and Implications
Phase III of the program (2010 to the present).                       Substantial yield gaps remain in large parts
                                                                      of rainfed and irrigated areas, indicating
Phase I won high marks for innovation (Singh                          significant potential for further gains. While a
and Swanson 2006; Anderson 2007; Swanson,                             number of factors influence yield gaps, from
Hall, and Reddy 2008). Independent evaluations                        the technology perspective, the immediate
by the Indian Institute of Management showed                          priority for addressing “technology fatigue” is
that the pilot mobilized farming communities,                         to resolve the “extension problem.” Research
developed public-private partnerships,                                also has significant challenges to confront,
improved interaction among farmers and                                such as dedicating more effort to crops other
extension workers, increased productivity and                         than rice and wheat and leveraging the rapidly
farm incomes (compared to control groups),                            emerging private sector to tackle important
and facilitated the development of supply                             outstanding and emerging challenges. These
chains for a number of commodities.113 The                            research challenges include the multiple biotic
112	 ATMA was scaled up under the government’s Support to             and abiotic stresses prevalent in the LIS, a
     State Extension Programs for Extension Reforms.                  transforming agriculture and food sector, and
113	 Participatory planning, operational flexibility, smooth and
     timely flows of funds, training and capacity building support,   the potential impacts of a changing climate.
     and independent monitoring and evaluation were some of
     the reasons for the pilot’s success. See World Bank (2007c).     114	 See Babu et al. (2012).




                                                                                                                  97
Republic of India: Accelerating Agricultural Productivity Growth




Technology capital is a critical input to accelerate      private sector research. Human capacity for
agricultural productivity. Global experience              technological innovation is needed in both the
amply demonstrates that success is more than              private and public sectors. The urgency need
the sum of public funding. The quality of                 to resolve the crises in the State Agricultural
innovation and the capacity of institutions to            University System is laid out in the Bhubaneswar
reconfigure and reorient themselves to a rapidly          Declaration. Similarly, the actions needed
transforming and increasingly knowledge-                  to improve agricultural extension are well
intensive agriculture are critical elements of            known but need to be put into practice. These
success. In this regard, priorities for research          institutional changes require fundamental
are to implement the reforms proposed by                  reforms, and implementing them requires the
panels led by eminent scientists within India             same unwavering political will and commitment
and to remove the remaining constraints to                that ushered in the green revolution.

Box 4: ATMA—Steps toward demand-driven, pluralistic extension

On the positive side, IFPRI’s study of India’s agricultural extension program concluded that ATMA had
widened the range of extension activities and modalities that could be funded, enabling extension to respond
better to the demands of different stakeholders, although performance varied widely across states, districts,
and blocks. Convergence across programs (such as the National Food Security Mission, National Horticulture
Mission) had improved, although again it varied by state, with good examples emerging from Maharashtra.
Regular meetings at the district and block level between different line departments (agriculture, animal
husbandry, fisheries, and so on) improved working relationships across departments. Additional funding
from ATMA helped some departments to implement their extension activities. ATMA’s association with
Commodity Interest Groups (CIGs) helps them to access commercial bank finance, knowledge, and other
services such as marketing and inputs—all important elements of success in extension. ATMA has created a
constituency for its support at the ground level through Farmer Advisory Committees (FACs) and CIGs and to
some extent expanded the reach of public extension agencies to rural communities.
On the less positive side, although ATMA is increasingly recognized as a new demand-driven, multi-agency
approach to extension at the district level, that role is not so well articulated at the block and village levels.
The lack of skilled staff at the right time remains a constraint. Like other government agencies, ATMA
experiences funding delays, but agricultural programs must meet seasonal needs, and delays seriously affect
agricultural productivity and incomes. Set up as an independent agency, ATMA was not administratively
linked to any particular department, which was instrumental to its success in Phase I. Subsequently the
Department of Agriculture became the lead department for ATMA, causing other line departments to
perceive ATMA as a Department of Agriculture scheme rather than an equal partner. It may be time to
explore more neutral organizational options. At the very least, other departments could take the lead
role where warranted (for example, Horticulture or Animal Husbandry in districts where they represent
the primary source of agricultural livelihoods). Links with agricultural research must be institutionalized
rather than left to personal initiative, interest, and contacts, especially at the block level. Partnerships
in delivering extension services should be encouraged with NGOs and the private sector, to add to the
plurality of service providers and encourage specialization based on comparative advantages (for example,
in areas such as agricultural marketing, where extension staff may not be the best equipped for the job).
Finally, monitoring and evaluation must be extended to monitor outputs, outcomes, and impacts, which
will require careful selection of performance indicators and systems to ensure that data are collected,
verified, and analyzed at regular intervals.
Source: Adapted from Babu et al. 2012.




  98
                        Livestock Subsector:
10                      Opportunities for Action to
                        Improve Performance
                                             115




Introduction                                                         inter-state variation in consumption prevails as
                                                                     well (Gandhi and Zhou 2010).117
Rising income levels, population growth,
and urbanization are shifting India’s dietary
                                                                     Supply has not kept pace with demand. India
patterns away from carbohydrate-rich cereals
                                                                     has some of the largest animal populations
to foods richer in proteins and micronutrients,
                                                                     but lower per capita availability of milk and
as noted in Chapter 1. Consumption of food
                                                                     meat than all other regions of the world.
from animal sources has increased steadily in
                                                                     Consumption of eggs is much lower than the
rural and urban areas (the shift in favor of fruits
                                                                     Asian average on account of low productivity.
and vegetables is even stronger). Interestingly,
these shifts are more pronounced in rural                            Looking to the future, continued income
than in urban areas and among the urban and                          growth and demographic changes in India
rural poor (Table 17). Despite past growth,                          are expected to propel strong growth in
consumption levels remain substantially below                        demand for livestock products, presenting
those in developed countries.116 Considerable                        huge potential for production growth. The

Table 17: Rural and urban household expenditure patterns have shifted away from cereals
                                                                Rural expenditure (%)                 Urban expenditure (%)
                                                               1993–94       2009–10                 1993–94       2009–10
Share of food in total household expenditure                     63.2          53.6                    54.7          40.7
Allocation of food budget:
Cereals                                                          36.6              28.2                24.7                21.7
Milk and milk products                                           14.3              15.5                17.2                18.5
Meat, eggs, and fish                                              5.0               6.2                 5.9                 6.4
Total animal products                                            19.3              21.7                23.1                24.9
Other foods                                                      44.1              50.1                52.3                53.4
All foods                                                       100.0             100.0               100.0               100.0
Source: Birthal 2008 and NSSO survey, 2009–10.
                                                                     117	 Milk consumption in Haryana, Punjab, and Rajasthan (at
115	 This chapter is based on Leitch, Ahuja, and Jabbar (2013).           146, 134, and 108 liters per capita annually, respectively) is
116	 In India, only about 10 grams of protein is derived daily per        way above that of states like Manipur and West Bengal (2.5
     capita from livestock products, compared to 16.6 grams in            and 20.8 liters, respectively). Per capita meat consumption is
     Asia, 45.3 grams in the Americas, 47.7 grams in the European         highest among the northeastern states but still relatively low
     Union, and 56.8 grams in Oceania (FAO 2011).                         at 10.4–19.4 kilograms.




                                                                                                                                  99
Republic of India: Accelerating Agricultural Productivity Growth




            gricultural GDP: Share of livestock,
Figure 63: A                                                                            Other estimates suggest that livestock
           1950/51–20110/11 (in 2004–05 prices)                                         production has large multiplier effects in
35.0%                                                 900                               India. An additional US$ 1 spent on primary
                                                      800                               livestock production generates US$ 4.7 in
30.0%
                                                                                        national household income (through demand




                                                            Thousand Crores (2004-05)
                                                      700
25.0%                                                                                   and nonfarm linkages), compared to US$ 4.3
                                                      600
20.0%
                                                                                        in fruits and vegetables, US$ 3.6 in crops, and
                                                      500
                                                                                        US$ 2.9 each in manufacturing and services
15.0%                                                 400
                                                                                        (FAO 2011). Importantly, most of the livestock
                                                      300
10.0%                                                                                   in India is owned by marginal and small
                                                      200
                                                                                        farmers, indicating that livestock ownership
5.0%
                                                      100                               is more equitable than land ownership. About
0.0%                                                  0                                 65 percent of India’s rural farm households have
        1951
        1954
        1957
        1960
        1963
        1966
        1969
        1972
        1975
        1978
        1981
        1984
        1987
        1990
        1993
        1996
        1999
        2002
        2005
        2008
        2011


                                                                                        marginal land holdings of less than 1 hectare.
                                                                                        These same households own 20 percent of all
    Livestock   Agriculture   Livestock Share (Left Axis)
                                                                                        land but more than 50 percent of all cattle,
Source: Authors, using CSO National Accounts Data.                                      small ruminants, pigs, and poultry and about
                                                                                        45 percent of India’s buffaloes (Table 18).
livestock subsector has grown at twice the
rate of the food grain subsector, raising the                                           Livestock is an important source of employment
share of livestock in agricultural GDP from                                             for rural households and especially for
20 percent in TE 1981 to about 30 percent in                                            women. Mixed crop-livestock farming systems
TE 2010 (Figure 63). Livestock’s contribution                                           predominate in India, although some scaling up
to agricultural growth increased from 31                                                is starting to occur in the leading dairy states.
percent in 1992/93 to 36 percent in 2008/09.                                            Livestock are labor intensive, providing an
Milk accounted for two-thirds of the total                                              opportunity for unskilled labor in rural areas.
value of output in livestock, owing to past                                             Crop agriculture is seasonal and risky; adding
and continuing public investment in the dairy                                           livestock production helps to reduce rural
sector (see Annex 10, Table A.10.1).                                                    underemployment, particularly of family labor,
                                                                                        and provides more stable income throughout
Livestock income contributes significantly to                                           the year. A recent study estimated that the
both poverty reduction and economic growth.                                             production of 1,000 liters of milk alone on
Using the nationally representative survey data                                         a daily basis by small, medium, and large
on farm households in 2002/03 (NSSO 2005),                                              producers, respectively, creates 230, 97, and 25
Birthal and Negi (2012) find that the probability                                       jobs in India (Staal, Nin-Pratt, and Jabbar 2008).
of a household being poor decreases more with                                           The livestock subsector engaged 8.8 percent of
the rising proportion of per capita household                                           the agricultural labor force in 2005 (Table 19),
income derived from livestock and crops. Over                                           with women comprising between 70 and 80
and above the negative effect of log per capita                                         percent of the workforce. While 90 percent of
income, the marginal effect on poverty is –0.36                                         employment in primary production is in rural
for the share of income from livestock and                                              areas, a sizeable number of jobs in marketing
–0.25 for the share of income from crops.                                               and processing are in the urban areas. Animal



  100
                          Chapter 10: Livestock Subsector: Opportunities for Action to Improve Performance115




Table 18: Distribution of land and livestock holding by land holding size
                                 Marginal                    Small      Semi-medium          Medium          Large
                                   (<1ha)                   (1–2 ha)      (2–4 ha)          (4–10 ha)      (>10 ha)          All
 Percent households                 64.8                      18.5          10.9               4.9            0.9           100
 Average land holding size (ha)     0.38                      1.38          2.68              5.74         17.08            1.23
 Proportion of land owned (%)       20.2                      20.9          23.9              23.1           11.8           100
 Proportion of livestock ownership (%)
 Cattle                             50.4                     22.2             16.1             9.1            2.2            100
 Buffalo                            44.8                     21.5             18.3            12.5            2.9            100
 Small ruminants                    52.4                     21.3             14.6             8.4            3.3            100
 Pig                                55.6                     22.6             13.6             6.5            1.7            100
 Poultry                            64.2                     18.7             11.2             4.1            1.8            100
Source: www.indiastat.com.
Note: Land ownership is for 2005–06 and livestock ownership is for 2006–07.


Table 19: Employment in the livestock subsector by farm size category
                      Agricultural employment in                 Share of livestock in            Share of women in livestock
                      rural employment (percent)             agricultural employment (%)                employment (%)
 Farm category          1993–94       2004–05                  1993–94          2004–05             1993–94       2004–05
 Landless                 62.8          62.5                      5.5             2.3                 68.0           97.2
 Marginal                 73.0          65.2                      7.1             9.2                 69.4           73.4
 Small                    89.4          88.2                      6.1             7.4                 72.1           82.1
 Medium                   92.2          90.8                      6.8             7.8                 72.8           83.1
 Large                    93.1          91.4                      7.7             8.6                 76.7           82.0
 All                      78.4          72.7                      6.8             8.8                 70.5          76.6
Source: NSSO 2006 data.


husbandry is also more inclusive, with 69                              India’s productivity is well below the world
percent of the persons employed belonging to                           average. Given the importance of livestock
SC, ST, and OBC.                                                       among poor households (for income and for
                                                                       coping with risk), growth in the subsector is
Finally, improving productivity in the livestock                       likely to continue.119 To avoid the tradeoff with
subsector is also important for mitigating                             environmental impacts, it is critical to focus
greenhouse gas emissions.118 With the exception                        on livestock management and productivity.
of dairy, past growth in the sector has been                           Growth in greenhouse gas emissions is closely
driven more by expanding numbers of animals                            related to the nature of the livestock production
than by improved animal productivity (output                           system. Emissions per head of cattle are
per unit of livestock or yields). Even in dairy,
                                                                       119	 Livestock growth is much less volatile than crop growth,
                                                                            and livestock have been a key part of household risk
118	 FAO (2006) estimates that about 18 percent of global                   management and food security strategies to cushion weather
     greenhouse gas emissions are from livestock.                           and other shocks (Annex 10, Figure A.10.1).




                                                                                                                                101
Republic of India: Accelerating Agricultural Productivity Growth




strongly dependent on diet quality (Harper                          by the lack of formal marketing channels, as
et al. 1999; FAO 2006). The following sections                      profit margins are lower for milk sold through
discuss more specific opportunities, constraints                    informal markets. During the flush production
to growth, and areas for action to improve the                      season, informal traders often offer less than
performance of the livestock subsector.                             market prices or even decline to procure some
                                                                    of the output.

Opportunities for the Livestock                                     Historically, the increase in average milk
Subsector                                                           yields was driven by better market access, the
                                                                    growing population of crossbred animals as
India is a global leader for livestock production,
                                                                    artificial insemination (AI) services expanded,
although production per animal remains
                                                                    and the replacement of draft breeds with
low by world standards.120 The average milk
                                                                    dairy animals (as mechanization spread).
yield of Indian cows is about 3.4 kilograms
                                                                    Growth in meat output has mainly come at
per day, compared to the world average of 6.3
                                                                    the extensive margin—that is, from growth in
kilograms per day. Unlike other countries,
                                                                    livestock numbers. All types of meat output
in India buffaloes are a significant source of
                                                                    increased (at varying rates) except for pig
milk, accounting for over half of national milk
                                                                    meat. Precise and systematic data on measures
production. Average milk productivity per
                                                                    of meat productivity in India are scarce, but
animal grew at a respectable pace from 1980 to
                                                                    the limited data available suggest a near
2000 but has slowed considerably since (Annex
                                                                    stagnation of productivity (measured crudely
10, Table A.10.2). Reasons for slower growth
                                                                    as carcass weight per animal) over the last four
in milk production include indiscriminate
                                                                    to five decades. The only segments that have
crossbreeding, the failure to respond to growing
                                                                    perhaps shown modest growth in productivity
feed scarcity, underinvestment in systematic
                                                                    (measured by feed conversion ratios) are the
disease prevention and control programs, and
                                                                    commercial broiler and egg segments.
inadequate services and advice to farmers
(Leitch, Ahuja, and Jabbar 2013).
                                                                    India has substantial room to increase output
                                                                    of milk and meat at the intensive margin. The
The formal processing sector expanded
                                                                    potential for milk production is demonstrated
rapidly, driven by the private sector after the
                                                                    in Figure 64, which shows the large variation in
deregulation of the dairy industry in the early
                                                                    milk productivity across states.121
1990s, but it remains small. Currently the
formal sector handles less than 20 percent
                                                                    Crossbred cattle are an important contributor
of India’s milk output. More than half of the
                                                                    to yield differentials, which in turn reflect the
registered milk-processing capacity is in the
                                                                    wide variability in AI dissemination rates across
private sector (in relation to the cooperative
                                                                    states. States with high rate of AI dissemination
sector). Small producers are particularly affected
                                                                    and other policy measures (such as the

120	 India has more than half the world’s buffalos, one-fifth of
     the world’s small ruminants, and more than 10 percent of       121	 Since animal productivity is relatively less subject to agro-
     the cattle population. With an estimated 127 million tonnes         climatic conditions, the differences across states suggest
     of milk in 2011-12, India leads the world in milk production        the presence of substantial “yield gaps” and scope for
     and has emerged as a leading meat exporter.                         improvement in many, especially the lower-income, states.




  102
                                                                Chapter 10: Livestock Subsector: Opportunities for Action to Improve Performance115




Figure 64: Milk yield variability by state
                                               60000
   Productivity per animal in milk (Rs/year)




                                                                                                                                                                          Milk
                                               50000

                                               40000

                                               30000

                                               20000

                                               10000

                                                   0




                                                                                                                                                                                                                                                Meghalaya
                                                       MP

                                                            Maharashtra

                                                                          Assam

                                                                                  Chhattisgarh

                                                                                                 Tripura

                                                                                                           Karnataka

                                                                                                                       Orissa

                                                                                                                                UP

                                                                                                                                     HP

                                                                                                                                          Nagaland

                                                                                                                                                     Uttarakhand

                                                                                                                                                                   Jharkhand

                                                                                                                                                                               WB

                                                                                                                                                                                    Bihar

                                                                                                                                                                                            AP

                                                                                                                                                                                                 TN

                                                                                                                                                                                                      Rajasthan

                                                                                                                                                                                                                  Gujarat

                                                                                                                                                                                                                            Manipur

                                                                                                                                                                                                                                      Mizoram



                                                                                                                                                                                                                                                            Goa

                                                                                                                                                                                                                                                                  Haryana

                                                                                                                                                                                                                                                                            Kerala

                                                                                                                                                                                                                                                                                     Punjab
Source: Chand and Parappurathu 2011.


development of market infrastructure and                                                                                                                           productivity. Expanding the value of meat and
promotion of milk procurement) experienced                                                                                                                         live animal exports requires that India address
high productivity gains. Feed constraints and a                                                                                                                    the pressing feed problem, improve the sanitary
high morbidity/disease burden also contribute                                                                                                                      quality of finished output, invest in controlling
to lower yields in some states. Given their                                                                                                                        foot and mouth disease (FMD), and implement
low initial productivity, LIS can effectively                                                                                                                      an identification and traceability system.
exploit the livestock subsector’s potential for
growth. For example, Bihar and Odisha have
recently shown high growth rates, supported                                                                                                                        Constraints and Challenges
by policy initiatives and the expansion of dairy                                                                                                                   For India’s livestock subsector to fulfill its
cooperatives.                                                                                                                                                      potential, a series of specific constraints and
                                                                                                                                                                   challenges require attention. They range
New opportunities for income growth are                                                                                                                            from the need to improve public spending on
emerging in export markets. India has already                                                                                                                      livestock to particular concerns surrounding
become one of the world’s leading exporters                                                                                                                        animal breeding and genetics, feeding and
of bovine (buffalo) meat. India is unlikely to                                                                                                                     nutrition, the environmental impacts of the
have a substantial dairy surplus for export, but                                                                                                                   subsector, animal health, markets and value
meat exports have significant potential to grow.                                                                                                                   chains, and financial and advisory services
India is not a competitive exporter of chicken                                                                                                                     related to livestock production.
and goat meat but is highly competitive in
buffalo meat, mutton, and pork. Expanding
                                                                                                                                                                   Public expenditure is low, and priority
markets for processed meat and halal-certified
                                                                                                                                                                   setting is needed
products provide additional opportunities for
export growth. Meat exports can be a means                                                                                                                         National expenditure on livestock is low,
of increasing the offtake rate to improve                                                                                                                          declining, and insufficient to exploit the



                                                                                                                                                                                                                                                                                        103
Republic of India: Accelerating Agricultural Productivity Growth




Table 20: Public spending in the livestock subsector in India, selected years
                                                        TE 1992-93        TE 2000-01        TE 2008-09
Public spending as % of total agriculture spending         13.6               9.9               4.6
Public spending as % of livestock output value              3.6               2.8               2.3
Composition of public spending (%)
Dairy development                                          41.5                 38.6           25.0
Veterinary services and animal health                      23.7                 24.1           29.1
Cattle and buffalo development                             14.0                 11.7           10.5
Sheep and wool development                                  2.7                  2.4            2.0
Piggery development                                         1.8                  0.5            0.4
Poultry development                                         3.1                  2.4            2.4
Fodder development                                          0.9                  1.0            1.0
Direction and administration                                4.2                  8.7           19.1
Research, education, and extension                          2.2                  3.0            3.0
Others                                                      5.8                  7.6            7.5
Source: Birthal and Negi 2012.
Note: TE = triennium ending.


sector’s contribution and potential. India’s          poor quality of feed is a major constraint for
public expenditure on livestock as a percentage       productivity growth and has implications for
of total expenditure in agriculture declined          the environment, but feed is a very low priority
from 13.6 percent in 1992/93 to 4.6 percent in        in budget allocations, along with research,
2008/09, even as the subsector’s contribution         education, and extension. The importance
increased from 22.8 percent of agricultural           of extension cannot be overemphasized;
GDP to 24.8 percent. Spending as a percentage         inadequate knowledge and management
of livestock GDP has declined steadily from           practices directly affect animal health and
3.6 to 2.3 percent over the same period               productivity.
(Birthal and Negi 2012). The composition
of the expenditure is also shifting toward            It is critical to prioritize and rationalize public
“direction and administration” at the expense         expenditures, especially in the provision
of productivity-enhancing activities, with the        of public versus private goods. Monitoring,
bulk of expenditures going to the salaries and        evaluation, and impact assessment of all
administration (Table 20). Notably, allocations       government livestock programs is urgently
for the smaller species—small ruminants,              needed to inform future programs, rationalize
poultry, and pigs—that yield more benefits for        resource allocations, and enhance the delivery
smallholders remain low and have declined             of services.
over time.
                                                      Breeding and genetics: Room for
The National Dairy Plan has a long-term
breeding policy and strategy in place, but
                                                      improvement
other segments of the subsector still have no         Breed improvement programs have generally
apparent policy or strategy. The scarcity and         been overlooked. India has a wide range of



  104
                         Chapter 10: Livestock Subsector: Opportunities for Action to Improve Performance115




animal genetic resources, but only 20 percent                  to be indiscriminate, with little attention to
are defined breeds. Productivity increases in                  breeding policy). The review finds mixed results
advanced dairy countries are mostly attributed                 for other species.
to genetic improvement (70–75 percent),
with the rest attributed to improvements in                    Quality breeding stock is limited, but
animals’ environment (USDA 2012; F. Miglior,                   constraints related to the available technologies
Canadian Dairy Network, 2012, personal                         (such as AI) lead to generally poor results. Some
communication). India embarked on breed                        success has been achieved through AI in the
improvement programs in the late 1990s, but                    leading dairy states (Andhra Pradesh, Karnataka,
less than 10 percent of bulls used for AI have                 Kerala, Punjab, and Tamil Nadu). Problems
come through any sort of genetic improvement                   included low conception rates (35–40 percent
program. The lack of good breeding stock                       on average), with NGOs and private AI centers
is a major obstacle for cattle but even more                   achieving higher rates than government-run
so for small ruminants and pigs. Breeding                      centers because the cows need to be brought to
strategies and plans need to be developed at                   the government location for breeding. Studies
the state level. Breed improvement schemes                     have shown that farmers are willing to pay
can be implemented and show results quite                      for better conception results, since the overall
rapidly, given the short-cycle nature of small                 cost per calf decreases with higher conception
ruminants, the relatively high heritability of                 rates. Low conception rates are a key factor
key performance traits such as average daily                   in reducing productivity, as they result in
weight gain, and the ability to select animals                 longer inter-calving periods and less milk
based on their performance.                                    production. High levels of infertility mean that
                                                               many cows do not produce milk at all, as they
India has largely relied on crossbreeding122                   never get bred. India has more than 6 million
to combine the adaptation of indigenous                        infertile cows.
breeds with the yield capacity possessed
by exotic breeds, but upgrading beyond                         For small ruminants, a recent review by
50–75 percent generally is impractical                         the South Asia Pro-Poor Livestock Policy
given India’s environmental and resource                       Programme, (SAPPLPP 2012a, 2012b) highlights
constraints. Even when practical, there are few
                                                               that the best-performing animals among local
examples of sustained genetic improvement
                                                               breeds can have great productive potential. The
with crossbreeds. Nimbkar and Kandasamy’s
                                                               screening and systematic selection of livestock
(2012) review finds stagnating performance of
                                                               populations could rapidly yield results to
crossbred cattle in farmers’ and institutional
                                                               benefit the poorer livestock-keepers who rely on
herds. Probable reasons for stagnation include
                                                               low-input systems.
the absence of progeny testing programs for
crossbred males and the use of unselected
crossbred males (in fact, crossbreeding tends                  Feeding and nutrition: Issues of scarcity
                                                               and quality
122	 Crossbreds, with varying level of exotic inheritance,     Poor nutrition is a leading production
     accounted for 24 percent of dairy cattle in 2007, but
     the proportion of crossbred small ruminants and pigs is
                                                               constraint, and all projections indicate severe
     negligible (GOI 2008b).                                   shortages of feed and fodder in the future



                                                                                                           105
Republic of India: Accelerating Agricultural Productivity Growth




(Ramachandra et al. 2007, Herrero et al. 2009).123                    residues for feed (Ramachandra et al. 2007) and
Feed costs are already rising as the costs of                         the reduced availability of high-quality residues
crop production increase and the available land                       like sorghum and millet, the need for better
decreases. For commercial poultry and dairy                           integration of crop and livestock production
producers, prices of feed such as sorghum and                         systems is apparent.125 Feed storage receives
maize have increased faster in recent years than                      little attention, despite the routine occurrence
the prices of milk, eggs, or meat. Between 1990–                      of feed shortages during the summer months.
91 and 2005–06, the chicken to maize price
ratio declined to half, and the egg to maize                          Poor integration of crop and livestock
price ratio fell from 0.33 to 0.26 (Blümmel and                       production also has consequences for crops and
Rao 2008).                                                            the environment. Manure is the second most
                                                                      valuable livestock product, yet improper storage
Access to quality seed of fodder crops remains                        and application limit its contribution to soil
a constraint: Only about 7 percent of the                             fertility or as an economic option for energy
estimated 10.5 million hectares of fodder                             and biofuels. A considerable proportion of crop
grown in India is planted with improved seed.                         residue is burned in Punjab and Haryana—
Commercially processed feed of assured quality,                       squandering an otherwise useful feed resource.
accessibly priced and accurately labeled, is in
short supply. For that reason, a large proportion                     Common property resources are diminishing
of feed concentrates are home-made, often with                        and their quality is declining. Pastoralists and
limited or no understanding of the nutritive                          the landless rely heavily on public grazing
value of the components or the nutritional                            land and common property to sustain their
requirements of the animals. This practice                            animals, but both poor and rich producers
increases wastage as well as producers’ costs.124                     relied equally on common property for feed
                                                                      (GOI 1999; Narain, Gupta, and Veld 2005).
Smallholders use crops for food and feed.                             Communal land declined by an estimated 14
In maize, sorghum, millet, and groundnut,                             percent between 1980/81 and 2008/09 and
for example, crop breeding can improve the                            now accounts for a mere 3 percent of India’s
quality of stover and other crop by-products for                      geographical area (Birthal and Negi 2012). These
livestock in addition to increasing grain yield,                      lands are degraded through overstocking and
but such opportunities are often overlooked.                          poor management or are increasingly lost to
Given the likely increase in reliance on crop                         encroachment. The management of common
                                                                      property requires alternative institutional
123	 Dikshit and Birthal (2010) predict that by 2020 India will       arrangements to reconcile administrative and
     need an additional 100 million tons of green fodder, 60          legal procedures and alternative management
     million tons of dry matter, and 9 million tons of concentrates
     for feeding its livestock population, excluding poultry.         practices to reduce the stocking rate, increase
     Poultry production would demand an additional 27 million         offtake through culling, and improve the land’s
     tons of feed (Robinson and Makkar 2012), which translates
     into an additional protein requirement of approximately
                                                                      silvopasture potential through regrowth and
     6 million tons (equivalent to 60 million tons of cereals or      management.
     2.4 million tons of soybean; see Makkar 2012).
124	 Garg, Biradar, and Kannan (2009) find that nearly 90 percent
     of cows in Gujarat were overfed, but with balanced rations,      125	 Lower-quality residues such as rice and wheat, unless
     costs decreased by 10 percent and production increased by             inexpensively treated with urea, have little nutritive value,
     10 percent.                                                           yet they are the main feed source for most livestock.




  106
                          Chapter 10: Livestock Subsector: Opportunities for Action to Improve Performance115




Environmental aspects of increasing                                   to avoid increased soil erosion, down-slope
livestock productivity                                                sedimentation, reduced water infiltration and
                                                                      groundwater recharge, and ultimately reduced
As pressure on water and land resources rises                         pasture production (Sheehy et al. 1996).
throughout India, it will not be sustainable
to meet the increased demand for livestock
                                                                      Climate change and increased weather
products by expanding animal numbers. To
                                                                      variability will also affect livestock development
produce one liter of milk in Gujarat requires
                                                                      and require adaptation strategies. India’s
2,000–4,600 liters of water, mainly to grow feed
                                                                      production environment is characterized
(Singh 2004), against the world average of about
                                                                      by heat stress, drought, and flooding, all of
900 liters of water per liter of milk. India’s
                                                                      which are likely to become more frequent.
water use per liter of milk exceeds the world
                                                                      A large number of adaptation strategies
average in most intensive and semi-intensive
                                                                      exist, but greater implementation is needed.
systems. For example, using several field
                                                                      Kurukulasuriya and Rosenthal (2003) provide
locations in the Indo-Ganga Basin, Haileslassie
                                                                      a useful framework to consider several
et al. (2011) find that the water used to produce
                                                                      adaptation options. Micro-level adaptation options
a liter of milk ranged between 1,300 and more
                                                                      include diversification and intensification
than 3,000 liters. Smallholder production
                                                                      of crop and livestock production, pasture
systems based on crop residues make more
                                                                      management, irrigation of fodder crops,
efficient use of water (Peden, Tadesse, and
                                                                      planning breeding seasons to coincide with
Misra 2007), but continuing pressures for
                                                                      resources, fodder storage, and better timing
intensification are likely to exacerbate the
                                                                      and management of disease prevention efforts,
pressure on water resources unless specific
                                                                      given that climate accounts for 52–84 percent
ways to increase livestock water productivity
                                                                      of the variation in FMD incidence. Innovative
are encouraged. Potential strategies include
increased use of crop residues and by-products,                       mechanisms include weather-indexed insurance
managing the spatial distribution of feed                             schemes to better manage weather risks
resources in ways that match availability more                        and support increased technology adoption,
closely with demand, and enhancing animal                             remote banking to promote savings instead
productivity (Peden, Tadesse, and Misra 2007).                        of overstocking, and marketing livestock to
                                                                      limit risk. Institutional changes include support
India is already witnessing the pressures on                          for producer organizations (for inputs and
land from its growing human population                                markets), management of common property,
and increasing demands for alternative uses.                          and the development of fodder markets.
Livestock stocking rates are nearly 5–10                              Technological developments include dual-purpose
times the recommended levels in many                                  varieties suitable for feed and fodder, varieties
areas.126 Livestock need to be managed in                             bred for better-quality crop residue, agroforestry
ways that maintain vegetative groundcover                             varieties, improved animal health, low-cost
                                                                      animal shelters, and better-adapted species
126	 In rainfed areas, the present stocking rate is 1–5 tropical
                                                                      and breeds, among many others. FAO is
     livestock units (1 TLU = 250 kilograms, or roughly the size of   currently mobilizing a major global initiative
     one indigenous cow) per hectare against a recommended 1
     TLU, whereas in arid zones the stocking rates are 1–4 TLU per
                                                                      on sustainable livestock development, initially
     hectare against a recommendation of 0.2–0.4.                     focusing on enhancing resource-use efficiency



                                                                                                                  107
Republic of India: Accelerating Agricultural Productivity Growth




in livestock production, reducing pollution,                        diseases persist.128 No accurate data are available
and restoring grazing and pasture lands. India                      on the total financial and economic losses
can benefit significantly by partnering in this                     caused by animal diseases. Various estimates
global initiative through impact assessments,                       put the losses between Rs. 50 billion and Rs. 132
trade-off analyses, and networking in search                        billion annually (Chawla et al. 2004; Ahuja
of appropriate technologies and institutional                       et al. 2008). More recent estimates (Indian
models.                                                             Immunologicals 2012) are several times higher.

Deficiencies in animal health services                              To be effective, national programs for priority
and disease control                                                 diseases need to be more systematic and
                                                                    targeted. The government has identified FMD
Animal health and veterinary services account                       as the highest-priority disease and launched
for the highest share of public expenditure                         a national program. This is a very good step,
in the livestock sector, but service provision                      and similar programs could be undertaken for
is not very effective. Disease prevention and                       control of PPR,129 brucellosis, and other diseases.
control are the shared responsibility of central                    The FMD program is not without formidable
and state governments. Despite increased                            challenges. It has seen a dramatic increase in
public expenditures and an expansion in the                         coverage, yet it falls considerably short of the
number of veterinarians and paravets, disease                       target of 85 percent needed for effective control.
prevention and control remain inadequate.                           Additionally, FMD control is complicated by
Weak capacity for diagnosis and insufficient                        factors such as the many serotypes of the virus,
drugs for treatment undermine the quality                           the lack of sufficient vaccine and systematic
and efficiency of the services provided. Disease                    vaccination, unrestricted movements of
surveillance suffers from underreporting,                           livestock across state and national borders, and
untimely reporting, or failure to report                            symptom-free small ruminants with the disease,
important diseases. The reporting hierarchy,                        which often goes undetected. Infrastructure
complex reporting forms, and lack of facilities                     for disease surveillance and reporting needs
for quick communication are additional                              to be expanded as well as consolidated for
constraints (Ahuja et al. 2008; WGAHD 2012).                        functionality and efficiency to provide the
                                                                    foundation for a needs-based system to deliver
Increased public expenditures on animal                             animal health services. Legislation related to
health must be accompanied by reforms in                            the control of infectious and contagious animal
animal health services.127 The high incidence                       diseases should be updated and enforced
of some major diseases inflicts substantial and                     effectively, making disease reporting mandatory.
increasing economic losses on all livestock-
owning households, particularly poor and                            As with FMD, classical swine fever, and many
marginal households. India has been declared                        other priority diseases, vaccine supplies are
free of rinderpest, but a number of other
                                                                    128	 Such as FMD, black quarter, haemorrhagic septicemia, blue
                                                                         tongue, PPR, sheep and goat pox, and classic swine fever.
127	 Reformed health and extension delivery system are reviewed          Poultry are affected by Newcastle disease, infectious bursal
     in greater detail in Ahuja (2004); Ahuja, Morrenhof, and Sen        disease, chronic respiratory diseases, and H5N1 (highly
     (2003); Ahuja, Umali-Deininger, and de Haan (2003); Ahuja           pathogenic avian influenza).
     et al. (2003).                                                 129	 Peste des petits ruminants (ovine rinderpest).




  108
                   Chapter 10: Livestock Subsector: Opportunities for Action to Improve Performance115




inadequate. Problems in ensuring a cold chain       of charge or at nominal or subsidized cost,
heighten the need to develop heat-stable            yet most of this subsidy does not reach the
vaccines. Except for the FMD vaccine, most          intended beneficiaries. Often, dispensaries and
other vaccines are produced by the obsolete         service centers lack the budget to stock drugs.
state biological units, which do not have the       Centers may also charge a nominal fee for home
modern infrastructure or qualified staff to         visits, especially to cover transport costs. In
operate effectively or comply with Government       such cases, the basic drugs are still supposed to
Minimum Standards.                                  be provided free or at a nominal cost. In reality,
                                                    the free services provided by the veterinary
Distribution of curative healthcare facilities      centers are limited to prescriptions issued
and human resources across states is not always     by veterinarians. In the case of home visits,
congruent with distribution of livestock, and       in addition to the cost of medicine and the
these facilities are inadequately funded to         transport fee, government veterinarians charge
deliver effective services of adequate quality.     visit fees, which are not substantially different
Only 28 percent of livestock producers currently    from those charged by private veterinarians
use veterinary services (World Bank 2009).          (Ahuja et al. 2003; Ahuja, Rajasekhar and Raju
Government services are often overloaded, and       2008). There is a wide variation between states
the ratio of livestock to veterinarian is among     in terms of fees charged for home visits. In
the highest in some of the poorest states, such     some of the leading dairy states like Punjab and
as Bihar, Chhattisgarh, Jharkhand, Madhya           Haryana, over 70 percent of farm-gate services
Pradesh, and Rajasthan. The distribution of         are provided by public sector veterinarians on a
veterinary institutions also reflects a bias        “private contract” basis, at commercial rates—
toward large animals, as states with larger small   which the larger commercial producers are
ruminant populations have fewer veterinary          willing to pay, as they get satisfactory service.
institutions. State livestock departments
are overstretched, with responsibilities for        Rapidly growing markets and value
a number of services (many of which would
                                                    chains require better organization and
ideally be privatized). NGOs and the private
                                                    attention to quality standards
sector are active in service provision in
advanced dairy states where the public sector       Markets for all product groups are by and large
also has a strong presence—perhaps because          unorganized, traditional, and fragmented,
emerging commercial livestock producers             except for some portion of the dairy, poultry,
are more able and willing to pay for private        and egg markets. Marketed surplus of livestock
services. The ratio of veterinarians to animals     and livestock products, and producers’
could be improved by employing more paravets,       participation in product and input markets,
who could handle general clinical services and      vary considerably between the main product
enable vets to concentrate on more complicated      groups.
problems of diagnosis and treatment.
                                                    Dairy marketing. Traditional dairy marketing
Veterinary services provided in dispensaries        systems have been complemented by
and other centers are supposed to include           cooperatives since the early 1970s, and the
prescription and basic drugs and vaccines free      private sector has been playing an increasingly



                                                                                                109
Republic of India: Accelerating Agricultural Productivity Growth




important role since the market liberalization        industry, especially as demands for quality and
of the early 1990s. Cooperatives dominated            safety are increasing in urban areas and meat
the dairy processing industry due to regulatory       exports are growing as well. The total meat-
restrictions on private processing firms. With        processing capacity in India is over 1 million
the removal of those restrictions, the structure      tonnes per year, of which about 40–50 percent
of dairy markets changed significantly. The           is utilized. Meat-processing infrastructure is
private sector quickly overtook the cooperatives      inadequate and of poor quality. In 2007–08, only
in numbers of processing plants and installed         29 percent of meat was produced in registered
capacity. Between 1996 and 2010, the share            slaughterhouses, with the rest coming from
of the private sector in the number of plants         unregistered slaughter slabs/points and markets
increased from 49 to 67 percent and in installed      (GOI 2008b; Birthal 2008). Adequate, standard
capacity from 44 to 58 percent.                       slaughter facilities are not available to provide
                                                      proper sanitary conditions. Slaughterhouses
Wide differences are seen between states in           also lack facilities for processing wastage and
processing capacity, the market shares held by        by-products, leading to a large loss of value.
private, cooperative, and traditional processors,     Quality and hygiene levels in the wet markets
and the disposal pattern of marketed milk.            are low because of improper handling and
Shares of national milk output and shares of          the lack of water, electricity, and facilities for
national processing capacity are converging in        hanging carcasses/flaying. These problems
the leading dairy states, although not in the         add to wastage and the deterioration of meat
lagging states. Uttar Pradesh and Maharashtra         quality.
have the highest concentration of processing
capacity (each state has 20 percent of national       India has fast emerged as a leading exporter
capacity); Gujarat has 11.2 percent; Tamil Nadu,      of buffalo meat, accounting for nearly one-
7.2 percent; and Haryana, 6.4 percent. Average        quarter of the world’s beef trade, compared to
plant size is fairly small in Uttar Pradesh (83,000   just 8 percent in 2009. Lower prices, the ability
liters per day) compared to Gujarat (1,667,000        to provide halal products, and robust demand
liters per day), Tamil Nadu (184,000 liters per       (largely from the Middle East, Africa, and
day ), and Haryana (168, 000 liters per day)          southern Asia) have contributed to the rapid
(Ahuja and Sharma 2005).                              rise of exports. In response to this rapid growth,
                                                      new slaughterhouses are emerging, providing
Meat marketing. The Indian meat industry is           farmers with new outlets for nonproductive
highly fragmented and unorganized. Offtake            buffalo heifers, bulls, and bull calves. India is
rates of live animals are low, and domestic           cost competitive in the export markets, but
markets for live animals and meat are few and         poor quality and food safety standards could
unorganized. Generally, market chains for cattle      block it from capitalizing on this potential.
and small ruminants are fairly similar, and each      As discussed in greater detail later, there is a
involves a number of intermediaries.                  critical need to invest in upgrading food safety
                                                      and quality standards throughout the value
Modern, sanitary slaughtering and optimum             chain, implement risk assessment mechanisms
utilization of by-products are the most               and infrastructure, and set up a nationwide
important issues to be addressed by the meat          animal identification and traceability system.



  110
                    Chapter 10: Livestock Subsector: Opportunities for Action to Improve Performance115




In the commercial poultry industry, the              the Food Safety and Standards Act 2006 was
dominant mode of production and marketing            promulgated. The Act consolidated previous
in the broiler industry is contract farming,         regulations and went into effect from August
while independent enterprises remain                 2011. The implementation of the law remains
dominant in the layer industry. Like dairy           very poor because laboratories are few in
production, commercial poultry production            number and lack trained personnel and
is also concentrated in a few states. For            infrastructure (WGAHD 2012). Quality and
example, in 2004–05, Tamil Nadu, Karnataka,          safety standards in domestic and export value
Maharashtra, and Andhra Pradesh produced 41          chains are managed through a number of
percent of total broiler output in the country,      regulations and implementing authorities with
and 78 percent of it was under contracts             little coordination. The main challenge here is
(Annex 10, Table A.10.3). Both the broiler and       to ensure the coordination and enforcement
layer industries have scaled up significantly,       of quality and safety standards along the value
including contract production units. The             chain, as effectiveness in individual nodes will
benefits to smallholders from the growth of          have a limited impact.
the sector, with its increased specialization and
economies of scale, have been limited to the         Very limited access to credit and
few states where the industry is concentrated.
                                                     insurance
The market is large and growing, offering            Access to credit and insurance coverage are
significant scope for other states, especially       limited, declining, and biased toward dairy
the lagging states, to participate. Creating the     and the leading states. Livestock’s share in
enabling environment to establish such value         agricultural output is consistently increasing, yet
chains or to create links to already functioning     its share in agricultural credit has dwindled over
value chains in other states would benefit           time (Annex 10 Table A.10.4). The preference
smallholders in the lagging states. Backyard         for lending to dairy and large animal producers
poultry is still a major source of poultry meat      reduces the chances for producers in lagging
and eggs in India, but little support is available   states to access credit. The institutional response
from the public or private sector to address the     for smallholders has been the microcredit
production and marketing problems of such            industry, with a growing number of schemes
small-scale producers.                               linking banks with Self Help Groups and private
                                                     banks entering the microcredit sector.
Quality and safety standards. Quality
and safety standards in all value chains for         Livestock insurance coverage has increased
livestock products are poor. As incomes and          from a relatively low base, but only 6 percent of
urbanization increase, so do concerns about          animals, mainly dairy animals, are covered. The
food standards, quality, safety, variety, and        figures are slightly higher for poultry because
convenience (Grace, Baker, and Randolph 2010).       of the Poultry Venture Capital scheme, in which
To address problems such as adulteration;            credit for commercial poultry production is
improper or overuse of antibiotics, hormones,        linked with insurance (WGAHD 2012). New
and additives; unhygienic handling of products;      insurance products generally target high-
and inappropriate transportation and storage,        yielding animals and commercial production



                                                                                                  111
Republic of India: Accelerating Agricultural Productivity Growth




systems. Private insurance providers are                      backed with appropriate policy and regulatory
working with pro-poor development agencies                    measures. The program emphasizes enhancing
in several states and have a sizeable number                  productivity and strengthening village-level
of clients, but they are encountering problems                institutions and milk collection systems.
of high transaction costs, improper selection                 Investments in breeding and feeding, as
of clients, and moral hazard in settlement of                 well as in strengthening and expanding
claims (Sirohi et al. 2008).                                  infrastructure for the production, processing,
                                                              and marketing of milk products, are targeted.
Livestock advisory and extension                              It proposes to develop new business models to
services are negligible                                       link the unorganized sector to value chains by
                                                              establishing new producer companies alongside
Access to advisory and extension services                     cooperatives to provide viable alternative forms
remains negligible for livestock keepers.                     of dairy marketing (GOI 2012).
Health service providers are also responsible
for providing extension services, but nationally
                                                              Lack of timely, accurate information on the
only about 4 percent of households claim to
                                                              livestock population is a fundamental problem.
have access to livestock-related information.130
                                                              Accurate population data and data at the animal
In contrast, 28 percent of farmers said they had
                                                              production level are lacking. The National
advisory support for crop production. Livestock
                                                              Dairy Plan aims to set up an information
producers tend to rely on the more progressive
                                                              network for animal productivity and health,
local livestock farmers and the media (TV, radio,
                                                              and the country’s first unique identification
newspapers) as their main source of livestock
                                                              and information database for dairy, called the
production information. In designing the
                                                              Information Network for Animal Production
content of extension packages and institutional
                                                              and Health.
arrangements for delivering extension services,
it is important to respond to market demands
so livestock producers can link to existing or                A similar approach would benefit the meat and
emerging value chains (Sirohi et al. 2008).                   other livestock subsectors, such as poultry and
                                                              pigs. With a dual production structure emerging
                                                              in the country, in addition to encouraging
Conclusions and Implications                                  commercial and large-scale operations,
                                                              special attention must be given to technology,
Policy interventions to raise productivity in
                                                              institutional, and policy measures that can
the livestock sector need to target service
                                                              support the development of smallholder
delivery and institutions for technology,
                                                              systems to produce animals more efficiently and
marketing, and animal health. The National
Dairy Plan provides an example of this kind                   sustainably, and compete profitably in a price-
of multi-pronged approach. It is a multi-state                and quality-conscious market.
initiative to improve animal productivity,
strengthen/expand infrastructure for milk                     Potential gains from genetic improvement
procurement at the village level, and enhance                 will be squandered if feed constraints persist.
milk processing capacity and marketing,                       The manufacture of concentrates is expected
                                                              to increase as livestock production scales up
130	 The rates are slightly higher in leading dairy states.   and becomes increasingly commercial, but


  112
                   Chapter 10: Livestock Subsector: Opportunities for Action to Improve Performance115




other avenues for improved feed supply will         Investments in market infrastructure are
require attention. These include green fodder       important for the efficient distribution of
production, dual-purpose food and feed crops        livestock products. A large proportion of
in farming systems, dual-purpose varieties for      the investment will necessarily come from
grain and stover, the treatment of crop residues,   the private sector. In addition, the public
facilitating feed market development, trade         sector needs to pay attention to measures
policy to guide domestic production versus          that will help smallholders improve their
imports of maize and soybeans, and better           bargaining power and build their capacity
management of common property resources.            to absorb production and market risks.
The priority given to these issues will vary        Options include policy and public support
depending on their relative importance in           for producer organizations, risk mitigation,
different states.                                   reputation-building through labeling or
                                                    branding programs, and improving access
Animals’ productive potential depends crucially     to information with respect to pricing and
on the animal health system, which has a poor       product quality.
performance record. The quality of livestock
support services remains inadequate and poor,       Improvements in extension services will be
and disease surveillance, control, diagnostics,     critical to promote modern practices to produce
and reporting continue to be weak. Serious          products that meet the quality requirements
steps are required to improve animal health         of domestic and international markets. An
and extension support systems (including            important part of these advisory services will
identification and traceability), identify areas    be monitoring and evaluating the activities of
of public and private investment, institute         animal health extension services provided by
policies attractive for the private sector, and     ATMA and KVKs.131 A strategy to develop the
reform public service delivery systems to           capacity of livestock extension staff is needed.
become more efficient and responsive. To ease       Programs to improve access to credit and
funding constraints, subsidies on drugs should      insurance services for smallholders, and for
be phased out, with government clinics either       animals other than the high-yielding
charging the full cost for essential drugs or not   larger animals (cattle), need to be developed
stocking them (Ahuja 2004).                         and piloted.




                                                    131	 Krishi Vigyan Kendras (Agricultural Science Centres) are
                                                         extension outreach programs.




                                                                                                               113
D. Moving Beyond the Farm: Investments,
   Markets and Food Processing




I
    ndia’s agricultural markets have long a long                     and food processing needed for the modern,
    history of active government intervention.                       diversified, and vibrant agriculture that India has
    The overall regulatory framework was initially                   tremendous potential to realize (Acharya et al.
established to prevent market failures (as in                        2012; Ganesh-Kumar et al. 2012; GOI 2013b).
the Bengal Famine) resulting from inadequate
state intervention (Acharya 2006). From the                          It is noteworthy that when and where
1960s, various instruments were introduced                           reforms are introduced, even if partially, the
to support the green revolution and achieve                          private sector response is swift and dynamic,
food security (Chand 2003b; Acharya 2004). The                       as witnessed by the emergence of contract
instruments introduced in the 1960s and 1970s                        farming, electronic exchanges, ICT-based
are now widely viewed as being outdated and                          market information systems and kiosks, and
unproductive, to the extent that the current                         myriad value chain improvements (Birthal et al.
situation is characterized a market failure “due to                  2006, 2012, Joshi; Gulati, and Cummings 2007;
excessive state intervention” (Acharya 2006).                        Gulati and Ganguly 2010; FICCI 2010; Readon
                                                                     and Minten 2013) and the recent rapid growth
Agricultural markets in India have been studied                      in exports (Gulati, Jain, and Hoda 2013).
extensively.132 The consensus from these
analyses is that India’s agricultural markets are                    Given the importance of marketing and other
inefficient and fragmented, lacking integration                      post-farm elements of agricultural value chains,
across parts of the country (Gulati, Landes,                         this section builds on the large knowledge base
and Ganguly 2009; Reardon and Minten 2013;                           to focus on three themes. The first is a brief
Sekhar 2012).                                                        review of the current situation with agricultural
                                                                     markets, especially the current status of the two
Some regulations have eased over time, especially                    main regulatory reforms—the APMC Act and
since the 1990s, but reform has been slow,                           ECA (arguably the two most binding constraints
uneven, and often marked with reversals. By and                      for the proper functioning of agricultural
large, the perception is that the reforms have                       markets). The second theme is a review of the
been insufficiently deep or broad to promote                         efficacy of alternative marketing arrangements
the level of marketing, trade, value addition,                       in reaching small-scale farmers. The third theme
                                                                     is the key issues associated with food processing,
132	 For an overview, see GOI (2007, 2012) and Reardon and
     Minten (2013); see also Rashid, Gulati, and Cummings (2008);    an understudied and underexploited avenue for
     Ganesh-Kumar, Gulati, and Cummings (2007); Mattoo, Mishra,      diversifying agriculture and increasing off-farm
     and Narain (2007); Joshi, Gulati, and Cummings (2007); Landes
     and Burfisher (2009); Gulati, Ganguly, and Shreedhar (2011);
                                                                     employment, particularly in the poorer and
     and World Bank (1999, 2005, 2007d, 2008b).                      more agriculturally dominated states.


                                                                                                                   115
11               Investments for Growth and
                 Sustainability




T
      he importance of investment for              participation is the way to the future. Yet
      productivity and growth cannot be            both public as well as private investments are
      understated. The relationships between       needed—they are not substitutes.
capital formation and agricultural growth and
labor productivity, as well as the “crowding       As a share of total capital formation in the
in” effects of public investment on private        economy, however, the trends continue to
investments, are well established (Larson et al.   decline for both public and private investments
2000; Stephan et al. 2009; FAO 2011; Bisaliah      (Figure 65). For the private sector, the trend
and Dev 2012; Chand and Parappurathu 2012).        reflects a revealed preference for investment
Not surprisingly, the “neglect” of agriculture     outside the sector, perhaps driven by higher
(GOI 2007) sparked serious concerns, given the     returns or by an inadequate enabling
lengthy decline in public investment from levels   environment for investment in agriculture.
that were already low—it peaked in the 1970s at    Public investment is slipping once again after
about 4 percent of agricultural GDP (Chand and     its brief increase in the mid-2000s.
Parappurathu 2012).
                                                              Share of agriculture in gross capital
                                                   Figure 65: 
As shown in Figure 10 earlier, public                         formation
investments increased after 2003/04. This major
reversal from the negative growth rates in the     30.00
1980s and 1990s (Bisaliah and Dev 2012) was the    25.00
outcome of a concerted effort through several
                                                   20.00
major programs (some in national “mission”
modes) during the Eleventh Five Year Plan, with    15.00
substantial budgetary resources allocated to       10.00
boost agricultural growth. Since 2006/07, the
                                                   5.00
trend has been relatively flat, however. Private
investment has consistently increased since        0.00
the mid-1990s, reaching about 20 percent of
                                                           1993-94
                                                           1994-95
                                                           1995-96
                                                           1996-97
                                                           1997-98
                                                           1998-99
                                                           1999-00
                                                           2000-01
                                                           2001-02
                                                           2002-03
                                                           2003-04
                                                           2004-05
                                                           2005-06
                                                           2006-07
                                                           2007-08
                                                           2008-09
                                                           2009-10
                                                           2010-11
                                                           2011-12




agricultural GDP, and now accounts for over
80 percent of total investment in the sector.                      Public        Private    Total GCF
This development is positive. Agriculture
is private enterprise, and increased private       Source: Authors, using CSO statistics.




                                                                                                    117
Republic of India: Accelerating Agricultural Productivity Growth




          Government expenditures on agriculture and allied sectors (all India, 2004/05 prices)
Table 21: 
 Particulars                                                                             TE 1995                   TE 2009
 Government expenditure for agriculture (Rs billions)                                       827                     2,202
 Share of capital expenditure in total agriicultural expenditure (%)                         17                        23
 Share of union government in total agricultural expenditure (%)                             26                        49
 Share of agricultural expenditure in AgGDP (%)                                              20                        35
 Share of agriculture in developmental expenditure (%)                                       23                        33
 Agricultural expenditure per capita of rural population (Rs)                             1,316                     2,966
 Agricultural expenditure per ha (NSA) (Rs)                                               5,794                    15,645
 Share of agricultural research and education in AgGDP (%)                                 0.45                      0.58
 Share of agricultural extension in AgGDP (%)                                              0.14                      0.15
Source: Singh 2011, using Combined Finance and Revenue Accounts, GOI (various issues).



Trends in Public Expenditure                                        Clearly, agricultural expenditures are
                                                                    substantial, now equivalent to one-third of
For a sector deemed to be a priority,                               agricultural GDP. The rising trends in intensity (as
agriculture’s continued diminishing share                           a share of sector GDP) and allocation (as a share
of public investment would seem to indicate                         of development expenditures) suggest that
otherwise. That assessment may not be                               agriculture is indeed receiving higher priority.
accurate, however. Using a broader definition of                    Another change over time is the significantly
public expenditure that includes expenditures                       larger role of the central (union) government
in the sector (directly though agriculture                          in funding the sector, financing almost half of
programs and institutions) and expenditures                         total spending in 2009.
for the sector (for complementary public
investments in other sectors like rural roads                       There is nevertheless an important distinction
and electricity, which also show the priority                       to be made between public expenditures and
that the government assigns to promoting                            public investments. Of total expenditures,
agricultural growth), Singh (2011) finds that                       capital expenditures reflect the expansion
public spending to support agriculture is                           of productive capacity or public investment,
substantially higher, in real absolute and                          and these have also risen, indicating a rising
relative terms, in 2009 (13 percent) than it was                    priority for capital formation (Figure 66).
in the mid-1990s (2.5 percent) (Table 21).133                       The broader measure of investment (capital
                                                                    expenditures directly in agriculture and
                                                                    indirectly for agriculture through other sectors)
133	 Singh (2011) includes a comprehensive assessment of total      is significantly higher than the CSO estimate.134
     central and state expenditures covering both the capital
     account (investment) and revenue account (noninvestment        The broader series also shows significantly
     or recurrent) expenditures incurred on rural infrastructure    faster growth than the CSO series, being
     (rural roads and electricity), irrigation (major and medium,
     minor and command area development), rural development
     and rural employment programs (including land reforms),
     expenditures on crop husbandry, soil and water conservation,   134	 The CSO data include largely major and minor irrigation
     animal and dairy husbandry, fisheries, agricultural research        capital expenditures, which is a much narrower definition
     and education, extension, and training.                             than that used by Singh (2011).




  118
                                                                                                                              Chapter 11: Investments for Growth and Sustainability




           Trends in public investment
Figure 66:                                                                                                                                  Subsidies and public investment as a
                                                                                                                                  Figure 67: 
           (2004/05 prices)                                                                                                                  share of agricultural GDP
          600                                                                                                                     30

          500                                                                                                                     25
                                                                   Broader series
          400                                                                                                                     20
Rs. billion




                                                                                                                                  15
          300
                                                                                                                                  10
          200
                                                                                                                                  5
          100                                                                                          CSO series
                                                                                                                                  0
                                                                                                                                          2007-08        2008-09         2009-10         2010-11
              0
                                                                                                                                       Public GCFA           Food Subsidy          Fertilizer Subsidy
                  1994
                         1995
                                1996
                                       1997
                                              1998
                                                     1999
                                                            2000
                                                                   2001
                                                                          2002
                                                                                 2003
                                                                                        2004
                                                                                               2005
                                                                                                      2006
                                                                                                             2007
                                                                                                                    2008
                                                                                                                           2009
                                                                                                                                       Other Ag. Subsidies   Irrigation Subsidy    Power Subsidy

                                                                                                                                  Source: 
                                                                                                                                          Based on data from Hoda and Gulati 2013 and the
Source: Singh 2011.                                                                                                                       Planning Commission website.


particularly high in the 2000s (15 percent).135 In                                                                                infrastructure (13 percent), consisting mainly
totality, however, investments still constitute                                                                                   of rural roads and rural electrification, rural
less than 23 percent of public expenditures,                                                                                      development (13 percent), and major and
indicating that the vast majority of public                                                                                       medium irrigation facilities (11 percent).
spending is not directed at increased public
investments.                                                                                                                      Singh’s analysis shows increasing investments
                                                                                                                                  as well as increasing concentration in
In addition to levels of expenditure, the                                                                                         delivery of private goods or subsidies. But
composition of expenditures is an important                                                                                       it does not fully account for the substantial
indicator of the relative priority placed on                                                                                      public expenditures on subsidies in various
different components of public expenditures.                                                                                      forms. Given the many concerns about the
A breakdown shows the largest share going                                                                                         composition of public expenditures and the
to rural infrastructure (34 percent), followed                                                                                    need to boost investments in agriculture
by input support services (26 percent), rural                                                                                     (Gulati and Narayanan 2003; Chand and Kumar
development (23 percent), irrigation (15                                                                                          2004; GOI 2007), a brief update on these other
percent) and agricultural research, education,                                                                                    budgetary expenditures is useful.
and extension (at under 2 percent). Among
these items, expenditures on input and                                                                                            In a separate detailed study, Hoda and Gulati
support services grew the fastest (at 22 percent                                                                                  (2013) estimate current (and recent) levels of
annually), which includes the provision of                                                                                        the major categories of subsidies supporting
private goods (including subsidies), followed by                                                                                  agriculture. These estimates also do not capture
135	 Singh (2011) includes agricultural input supply and services
                                                                                                                                  the totality of subsidies to agriculture, such as
     as part of the broader series to capture the expenditures that                                                               subsidized credit or subsidies provided through
     “influence agricultural growth.” Earlier analysts have made
     similar arguments in analyzing public expenditures, using
                                                                                                                                  various rural development or other programs
     varying definitions and time periods (for example, see Fan                                                                   that may be reflected in Singh’s analysis, but
     et al. 1997; Chand 2000; Roy and Pal 2002). How much of
     the capital expenditures estimated by Singh fall under the
                                                                                                                                  which likely account for some of the largest
     category of inputs and services is not specified in the paper.                                                               elements in the total.


                                                                                                                                                                                              119
Republic of India: Accelerating Agricultural Productivity Growth




Comparing the findings of Hoda and Gulati                           The second program, RKVY,138 is the
with the data on gross capital formation (CSO                       government’s flagship effort to consolidate
estimates) in Figure 67 shows that the share                        various programs and provide essentially
of five major subsidies in public expenditures                      unconditional transfers to states to flexibly
continues to far outweigh public investments.                       align public expenditures to their priorities. The
As noted, the CSO data are more limited,                            review of RKVY for the same states also shows a
but even when using an approximation of                             clear preference for distributing inputs in Bihar
the broader public investments from Singh’s                         (87 percent). The extent of subsidy provision is
(2011) estimates, which is roughly double                           less in Andhra Pradesh (69 percent) and Gujarat
the CSO estimate, subsidies still dwarf public                      (42 percent) but still substantial (Annex 11,
investments.136                                                     Table A.11.2). Gujarat has paid relatively more
                                                                    attention to delivering services and building
As discussed, to kick-start agricultural growth,                    institutions and capacity, showing selectivity
the government initiated a number of programs                       in targeting key constraints (Raturi 2011;
or “missions” targeting substantial resources                       Feroni and Zhou 2012). The other two states
to raise agricultural productivity or reduce                        have focused primarily on input distribution to
poverty, which may also be contributing to                          drive production. Overall, both the assessments
capital formation in agriculture (Dev 2013a).                       indicate the pervasiveness of subsidies in
Among these, the two largest schemes are the                        various programs and schemes designed to
National Food Security Mission (NFSM) (2007)                        promote agriculture.
and Rashtriya Krishi Vikas Yojana (RKVY)
(2007), providing funds equivalent to about
US$ 2 billion per year to state governments                         Trends in Private Investment
toward agricultural development.137 A review of                     The aggregate CSO data show that private
their implementation in three states (Andhra                        investment has increased significantly since
Pradesh, Bihar, and Gujarat, chosen to represent                    about 2000 (Figure 10).139 The vast majority (90
different levels of development) for 2013/14                        percent) of these investments are estimated
shows subsidies rather than capital formation to                    to be made by farmers for on-farm production
be the preferred mode to stimulate agriculture.                     (HLC 2009). Data from the district-level
Depending on the state, 88–98 percent of the                        database show trends in farm equipment
NFSM funds were allocated to the distribution                       that are consistent with the CSO investment
of subsidies for purchasing various inputs (see                     trends (Figure 68), with remarkable growth in
Annex 11, Table A.11.1).                                            irrigation pumps per hectare since about 1998.
                                                                    Machinery per hectare (defined as tractors and
                                                                    power tillers) also shows sustained growth at an
                                                                    impressive but relatively lower rate.
136	 This approximation overstates the share of pure public
     investments, as Singh’s (2011) analysis does not look at the
     composition of sectoral expenditures—for example, how          Evidence on farm investments at the household
     much of the expenditures on power are for subsidies or how
     much of the rural development expenditures are in the form     level is limited and somewhat mixed. Using
     of subsidized private goods.
137	 See http://rkvymis.dacnet.nic.in/                              138	 See http://rkvy.nic.in/.
     (S(aqx2bynqxqv2ctrtauu2gjt2))/Reports/                         139	 The CSO data are based on assumptions regarding credit
     SectorWiseApprovedCost.aspx, http://www.nfsm.gov.in/                growth that raise doubts about the actual magnitudes of
     ReleasedApproval.aspx.                                              investment in agriculture (Dev 2013a).




  120
                                                         Chapter 11: Investments for Growth and Sustainability




Figure 68: Annual growth in farm equipment                             Composition of household assets (other
                                                            Figure 69: 
                                                                       than real estate), real 1999 prices

70.0%                                                       100%
60.0%                                                        90%
50.0%                                                        80%
                                                                        42%            44%
40.0%                                                        70%
                                                                                                  Financial
30.0%                                                        60%
                                                                                                  Consumer Dur.
20.0%                                                        50%        20%            19%        Business
10.0%                                                        40%                                  Livestock
                                                                         2%            6%
 0.0%                                                        30%        12%                       Agric. Capital
                                                                                       12%
-10.0%                                                       20%
                                                                        24%            20%
         1990
         1991
         1992
         1993
         1994
         1995
         1996
         1997
         1998
         1999
         2000
         2001
         2002
         2003
         2004
         2005
         2006
         2007
                                                             10%
                                                              0%
                  Pumpsets/ha      Machinery/ha                        1999            2007

Source: Authors, using ICRISAT-NCAP district database.      Source: NCAER-REDS survey data.


the CoC data from the Ministry of Agriculture,              but the shares of real estate (land and buildings)
Bisaliah and Dev (2012) suggest that nonland                remained the same. Among other assets,
capital stock per hectare declined on average               the share of farm assets (including livestock
between 1994–95 and 2007–08. The data                       and physical agricultural capital) declined,
are limited and may not represent all forms                 while business capital expanded (Figure 69).
of farm capital. Bisaliah and Dev do find,                  Households are also saving more, increasing
however, that irrigation capital showed positive            their financial assets.
growth, whereas livestock (animal capital) had
a negative growth rate. Within the limited                  While the average value of land owned by
accounting for capital, they identify a shift in            households increased (5 percent) between the
the composition of nonland capital, with an                 two years, average farm size declined from
increase in the share of irrigation capital. They           5.61 acres to 4.46 acres (about 3 percent per
also find an increase in factor productivity for            year). Assessing trends in investment (change
land, labor, and capital, suggesting that perhaps           in capital stock) is thus more appropriately
the estimates of capital (excluding animals and             done on a per acre basis (capital–land ratio)
irrigation capital) may be not be accurate.                 to account for changes in farm sizes. Table 22
                                                            shows the levels and growth in the value
A more complete accounting for household                    per acre of different types of farm assets
assets, including farm capital, is available in the         that households possessed in 1999 and 2007.
NCAER-REDS household survey data for 1999                   The growth in mechanized assets (which
and 2007. Land constitutes an overwhelming                  includes irrigation equipment) and livestock
77 percent of all assets owned by households,               capital is lower than nonmechanized capital.
and buildings (residential or business) account             Nevertheless, the data suggest robust annual
for another 10 percent. Total assets grew by a              growth in all forms of farm capital, with the
little over 5 percent per year over the period,             most rapid growth in nonmechanized capital.



                                                                                                          121
Republic of India: Accelerating Agricultural Productivity Growth




Table 22: On-farm investment (capital per acre), real Rs. 1999
                                      1999                     2007              Annual growth rate
a. Farm (mechanized)               2,771.48                   5,552.91                   9.1%
b. Farm (nonmechanized)              489.72                   2,152.73                  20.3%
c. Farm (other)                      555.94                   1,121.46                   9.2%
d. Livestock                       4,289.62                   9,784.51                  10.9%
Agricultural capital (a–c)         3,817.14                   8,827.10                  11.0%
All farm assets (a–d)              8,106.76                  18,611.62                  10.9%
Source: NCAER-REDS survey data.


These trends also suggest that although the CSO       mid-1990s and recovered only slightly after
data on private investments may not reflect           2004/05. This limited expansion may reflect
a complete picture, their underlying rising           the constraints on expanding groundwater
trend is corroborated by micro-level estimates.       irrigation (Shah 2007), but most public
The divergent trend in the CoC data is a bit          expenditures on irrigation go to major and
worrisome, though it should be reiterated             medium-scale surface water irrigation facilities
that they do not represent the totality of farm       (Singh 2011). Despite the substantial potential
investments and hence may not capture some            for surface water expansion (GOI 2013a), this
of the dynamic shifts taking place. Both data         potential does not appear to be translating into
sets indicate that the level of irrigation has gone   additional irrigated area.
up. The REDS survey data show that irrigated
area saw strong growth (as a share of overall         In this context, it is important to note that the
area owned by households) and unirrigated area        geographical targeting of public expenditures
contracted.                                           remains skewed. Singh (2011) analyzes
                                                      geographical distribution of public expenditures
                                                      (broadly defined); Annex 11, Figure A.11.1
Impact of Expenditure Patterns:
                                                      summarizes the results. Over 2000–09, relative
Sustaining Productivity                               to other parts of India, the eastern states
Two major trends identified in this review of         (excluding the northeastern region) received the
public and private investments are the rapid          lowest level of public spending for irrigation as
growth in irrigation and the large and growing        well as infrastructure, both on a per capita and
volume of subsidies in public expenditures. The       per hectare basis—a continuation of historically
importance of irrigation in India’s agricultural      skewed public expenditure patterns. This low
growth is clear from previous chapters and            level of expenditure persists even though
encapsulated in the close relationship between        eastern India has significantly less irrigation
irrigation and yields (at the aggregate level)        coverage (with the notable exception of Bihar)
shown in Figure 70 (left panel). The analysis in      than the average for all India (45 percent)
this chapter indicates that public expenditures       (Annex 11, Figure A.11.2).
on irrigation have grown rapidly, but their
impact on irrigated area is not evident. In           The main driver of irrigation in India over the
fact, growth in irrigated area slowed in the          past three decades has been private investment



  122
                                                                                                                                                                                     Chapter 11: Investments for Growth and Sustainability




           Trends in share of irrigated area and yields (left panel) and decadal trend growth rates in
Figure 70: 
           share of irrigated area (right panel)
55                                                                                                                                                                             2000     3.5
                                                                                                                                                                               1800
                                                                                                                                                                                          3
45
                                                                                                                                                                               1600
                                                                                                                                                                               1400     2.5
35
                                                                                                                                                                               1200       2
25                                                                                                                                                                             1000
                                                                                                                                                                                        1.5
                                                                                                                                                                               800
15                                                                                                                                                                                        1
                                                                                                                                                                               600
                                                                                                                                                                               400
 5                                                                                                                                                                                      0.5
 0                                                                                                                                                                             200
                                                                                                                                                                                          0
-5                                                                                                                                                                             0



                                                                                                                                                                                              1960-61
                                                                                                                                                                                                        1963-64
                                                                                                                                                                                                                  1966-67
                                                                                                                                                                                                                            1969-70
                                                                                                                                                                                                                                      1972-73
                                                                                                                                                                                                                                                1975-76
                                                                                                                                                                                                                                                          1978-79
                                                                                                                                                                                                                                                                    1981-82
                                                                                                                                                                                                                                                                              1984-85
                                                                                                                                                                                                                                                                                        1987-88
                                                                                                                                                                                                                                                                                                  1990-91
                                                                                                                                                                                                                                                                                                            1993-94
                                                                                                                                                                                                                                                                                                                      1996-97
                                                                                                                                                                                                                                                                                                                                1999-00
                                                                                                                                                                                                                                                                                                                                          2002-03
                                                                                                                                                                                                                                                                                                                                                    2005-06
                                                                                                                                                                                                                                                                                                                                                              2008-09
     1960-61
               1963-64
                         1966-67
                                   1969-70
                                             1972-73
                                                       1975-76
                                                                 1978-79
                                                                           1981-82
                                                                                     1984-85
                                                                                               1987-88
                                                                                                         1990-91
                                                                                                                   1993-94
                                                                                                                             1996-97
                                                                                                                                       1999-00
                                                                                                                                                 2002-03
                                                                                                                                                           2005-06
                                                                                                                                                                     2008-09




                                                       Area Irrigated (%)                                          Yields                                                                                                    Net Irrigated Area                                           Gross Irrigated Area

Source: Authors, using DES, MOA data.


in pumps. The International Water Management                                                                                                                                            Not only are the marginal returns to subsidies
Institute estimates that almost 62 percent of                                                                                                                                           significantly lower than those for other
irrigated area is under groundwater irrigation.                                                                                                                                         investments—particularly rural roads, research,
Power and credit subsidies, combined with                                                                                                                                               extension, and education—but they are also
heavy investment for rural electrification,                                                                                                                                             likely to be negative when the impacts of
have led to a proliferation of electric pumps                                                                                                                                           subsidies on environmental degradation are
for groundwater irrigation. Irrigation has                                                                                                                                              considered. This negative impact may already
been and will remain an important driver of                                                                                                                                             be reflected in slowing productivity and
agricultural productivity—but the main concern                                                                                                                                          yield growth despite farmers’ increasing use
is its sustainability. It is widely recognized that                                                                                                                                     of inputs. Specifically, the sustainability of
the distorted incentives created by subsidized                                                                                                                                          agricultural production itself is threatened as
pricing of electricity and credit have encouraged                                                                                                                                       excessive extraction of groundwater makes
the excessive extraction and highly inefficient                                                                                                                                         less water available for future use and water
use of water. On the benefits side, as shown by                                                                                                                                         levels fall below a critical threshold from which
Fan, Gulati, and Thorat. (2008) and summarized                                                                                                                                          they may not be able to recover. The political
in Annex 11, Figure A.11.3, power and credit                                                                                                                                            economy of subsidies in India is complex
subsidies had large marginal returns in terms                                                                                                                                           (Birner, Gupta, and Sharma 2011). The public
of improved productivity and poverty reduction                                                                                                                                          expenditure efficiency as well as budgetary
in the 1960s and 1970s. The marginal returns                                                                                                                                            implications have been relegated to the extreme
are now greatly diminished and are significantly                                                                                                                                        margins of public decision making. Yet the
less than returns to investments in roads,                                                                                                                                              more concrete evidence that is presently
research, and education. The returns to other                                                                                                                                           emerging on the outsized impacts that current
subsidies are also marginal.                                                                                                                                                            policies will have over the long run calls for



                                                                                                                                                                                                                                                                                                                                                    123
Republic of India: Accelerating Agricultural Productivity Growth




           Virtual water imports and public
Figure 71:                                                          6.7 percent reduction in groundwater extraction.
           procurement of rice                                       A 10 percent reduction in subsidy also leads
25                                                             0.8   to a reduction in the value of water-intensive
20                                                             0.7   agricultural output by 3 percent, largely due to
15
                                                               0.6
                                                                     the change in the value of rice. These analysts
10                                                                   also formally show that electricity subsidies
 5                                                             0.5
                                                                     cause an increase in the area planted to water-
 0                                                             0.4
                                                                     intensive crops such as rice.
 -5                                                            0.3
-10
-15
                                                               0.2   The long-term environmental cost and potential
-20                                                            0.1   threat to the sustainability of food production
-25                                                            0     can be derived from the findings of a separate
                                                                     study on the impact of groundwater depletion
            Punjab
                UP
           Haryana
                MP
                 AP
         Rajasthan
       Tamil Nadu
      Chhattisgarh
          Karntaka
      West Bengal
                HP
             Assam
            Gujarat
      Maharashtra
             Orissa
        Jharkhand
             Kerala
              Bihar

                                                                     on agriculture. Sekhri (2013) estimates, also
                                                                     using district groundwater panel data combined
                                                                     with production data, that a fall in groundwater
      Net Water Import (1997-2001, left axis)                        level by 1 meter reduces food grain production
      Share of Production Procured by FCI (1997-2001, right axis)    by 8 percent, water-intensive crops by 9 percent,
      Linear trend of share of production procured by FCI            and cash crops by 5 percent.
Source: Kampman 2007 and Authors using DES, MOA data.
                                                                     Together with MSPs, these policies have driven
urgent attention—not only in the interest of                         the changes in crop composition evident in
sustainability but in the interest of responding                     several parts of the country, particularly in the
to the serious risks imposed by climate change.                      Northwest and Mid-West, where farmers have
The emerging trends and risks are discussed                          concentrated on water-intensive crops such as
next in greater detail.                                              rice and sugarcane. These areas are experiencing
                                                                     the most severe groundwater crises. As shown in
                                                                     Annex 11, Figure A.11.4, the most irrigated crops
The energy-water-agriculture nexus
                                                                     are cereals and sugarcane, and not surprisingly
The negative impacts of free or highly                               they are cultivated in the areas with critical
subsidized power supply, including falling                           groundwater levels.141 This correlation is shown
groundwater levels and problems with soil                            more clearly in Figure 71, which summarizes
and water quality resulting from overuse of                          virtual water flows across states in India. The
groundwater, have long been debated. Badiani                         figure also shows the strong correlation between
and Jessoe (2011) provide rigorous evidence on
                                                                     141	 Less than 60 percent of the area under rice is irrigated,
the impact of electricity subsidies on agriculture                        largely because agro-climatic conditions favor rice in
as well as the extraction and overexploitation                            most parts of the country with less access to irrigation.
                                                                          Consequently, wheat has performed significantly better in
of groundwater.140 They find that a 10 percent                            terms of productivity growth. Sugarcane, on the other hand,
reduction in the average subsidy leads to a                               has not done so well. Considering that sugarcane had among
                                                                          the highest rates of area expansion in the recovery phase,
                                                                          and that virtually all of that area was irrigated (especially in
140	 They use district groundwater data along with electricity            the major growing states of Uttar Pradesh and Maharashtra),
     prices and generation data and agricultural production data          the low improvement in yields may reflect increasing water
     from the respective government ministries/agencies.                  stress with excess groundwater extraction.




  124
                                                  Chapter 11: Investments for Growth and Sustainability




water “export,” essentially groundwater in the                    Impact of nutrient imbalance on land
                                                       Figure 72: 
Northwest, and public procurement of rice.                        productivity
The virtual flow of water from the water-scarce                           40000
Northwest to the water-surplus East goes against
the intent of current development strategy,                               30000




                                                      Land Productivity
which is to promote the reverse flow.
                                                                          20000
Electricity is an important input to productivity
growth, but the new, concrete findings                                    10000

highlight the unsustainability of its current
                                                                              0
pricing policies: Electricity subsidies drive over-
                                                                                  0             5                10                    15
exploitation of groundwater, resulting in a rapid
                                                                                           Ratio of N to P Consumption
fall in groundwater levels, which in turn drives
                                                                                      VOP per ha TE1991        VOP per ha TE2008
down agricultural production and productivity.
                                                                     Vertical lines are TE2008 Medians for (L to R): India, PJ, HR, BH
Given the fact that water levels have already
fallen rapidly, it is likely that recent growth in     Source: Authors, using ICRISAT-NCAP district database.
agricultural productivity has been lower than it
could have been otherwise.                             impact of the ratio of nitrogen to other
                                                       fertilizers on TFP. Using district-level data, a
Consequences of imbalanced                             simple nonparametric relationship between
                                                       land productivity (defined as value of total
nutrient use
                                                       output per hectare) and the ratio of nitrogen to
Another potential policy-driven factor                 other nutrients provides a clearer visualization
suppressing productivity is the nutrient               of the likely magnitude of this impact. Figure 72
imbalance resulting from current patterns of           shows the change in land productivity as the
fertilizer use. The environmental impacts of           ratio of nitrogen to other nutrients rises. The
imbalanced soil nutrients are widely recognized        response initially rises and then falls as the
(see, for example, GOI 2007). When farmers             ratio increases.
apply significantly more urea than the other
major nutrients (potassium and phosphorous),           The drop is quite significant at high levels,
let alone micronutrients, they inadvertently           indicating a substantial cost in terms of
create soil and water problems and contribute          lower productivity. More importantly, the
to greenhouse gas emissions. Again, the                figure shows how the response function
major factor behind this practice is the price         has changed between 1991 and 2008. The
distortions created by substantial subsidies that      response at lower levels is higher, showing
alter the relative price of domestically produced      productivity growth (due to technical change
urea relative to the other nutrients.                  and irrigation expansion), but at higher levels,
                                                       the decline is sharper and eventually becomes
The direct impact of the imbalanced use                lower, indicating a long-term deterioration of
of fertilizers on productivity—that is, the            productivity.142
current as well as the long-term cost—is less
well appreciated. The econometric analysis in          142	 The response in rice and wheat yields is very similar, with the
Chapter 7 provided evidence of the negative                 main difference being a sharper fall at a higher nitrogen ratio.




                                                                                                                                 125
Republic of India: Accelerating Agricultural Productivity Growth




The figure also shows the all-India median         Conclusions and Implications
of the nitrogen ratio, which is still below
the turning point. For Punjab and Haryana,         Public spending has grown rapidly since
however, the median is now at the peak             the early 2000s and now equals one-third
response level, indicating that half of the        of agricultural value added, reflecting
districts are operating on the declining           the priority placed on agriculture by the
returns part of the response curve. This is not    government. But only a small fraction of this
surprising, given the long history of fertilizer   public expenditure goes toward expanding
use in these states. The surprising result shown   the productive capacity of agriculture (that is,
in the figure is the average for districts in      toward capital investments). Subsidies dwarf
Bihar: Even in this state, where productivity is   public investments. The debate on striking
relatively low, more than half of the districts    a better balance between the much higher
operate on the decreasing returns part of the      return investments and subsidies is not a new
response curve. In other words, the overuse of     one (see GOI 2012). The challenge has been
nitrogen is hurting rather than helping Bihar’s    the political economy of subsidies, which has
farmers, whose productivity is already very low.   made their budgetary implications seemingly
                                                   inconsequential in public policy decisions.
These results clearly indicate that policies       But evidence is starting to show that these
intended to help increase production and           policies are now imposing substantial costs.
productivity are in fact imposing substantial      If public policies are not rationalized, they
costs. More importantly, the costs are being       will further reduce productivity, compromise
borne right now. If they are not checked,          sustainability and lower the returns to
they are only likely to rise, resulting in lower   farmers—outcomes that are diametrically
productivity and returns to farmers.               opposite to their intent.




  126
12                     Current State of Agricultural
                       Markets in India




V
       irtually all agricultural produce is bought                Despite these problems, with rapidly rising
       and sold within India’s vast network                       incomes and urbanization changing the
       of market yards, subyards, and Rural                       patterns of demand, all “modern” market
Periodic Markets.143 The traditional marketing                    channels—private (modern retail, food
chain, passing through agricultural wholesale                     processing, the food service industry) as well as
markets and traditional urban retail, remains                     public (parastatals)—show high annual growth
dominant for agricultural commodities. Four                       rates. Modern retail is growing the fastest at
alternatives have emerged over time, based                        50 percent annually, although it has started
on distinct types of midstream or downstream                      from a very low base. Modern retail is estimated
buyers: modern retail/cooperatives, the                           to account for about 3–5 percent of the produce
processing sector, the food service sector,                       (nonstaple) market (Readon and Minten 2013).
and parastatal marketing (Reardon and                             The expansion of modern retail could have
Minten 2013).                                                     several important effects on the rural-urban
                                                                  food supply chain as experienced in many other
Research on India’s traditional markets                           countries (Reardon, Timmer and Minten 2010).
indicates that they are inefficient;144 lack                      It could change the processing sector and the
integration;145 are plagued by trader collusion;146               way it does business. It could reduce transaction
and are characterized by a high level of physical                 costs, make pricing more transparent, introduce
wastage.147 Wholesale market infrastructure                       economies of scale in transport, and enhance
for staple as well as nonstaple crops is not very                 food safety and quality. Urban consumers
developed.148 Agricultural markets suffer from                    could benefit from lower prices and lower
significant policy distortions, and progress                      transaction costs.
in implementing reforms has been slow and
uneven (Chand 2012).
                                                                  Even the traditional value chains for staples are
143	 There were 7,246 regulated markets at the end of 2011 and    seeing important changes (Reardon and Minten
     21,238 recognized Rural Periodic Markets, about 20 percent   2013). Upstream, farming technologies and
     of which worked under the ambit of market regulations
     (Patnaik and Sharma, 2013).
                                                                  farmers’ factor markets are quickly changing,
144	 Umali-Deininger and Deininger (2001); Ramaswami and          while in output markets, the majority of
     Balakrishnan 2002; Thomas (2003); Matoo, Mishra, and
     Narain (2007); Gulati, Landes, and Ganguly (2009).
                                                                  farmers are empowered by access to mobile
145	 Palaskas and Harriss-White (1996); Sekhar (2012).            phones. Midstream, fewer intermediaries seem
146	 Banerji and Meenakshi (2004).
147	 Matoo, Mishra, and Narain (2007).
                                                                  to be involved in operations (a process described
148	 Fafchamps, Vargas-Hill, and Minten (2008).                   as dis-intermediation) that are growing in scale.


                                                                                                             127
Republic of India: Accelerating Agricultural Productivity Growth




Downstream, large price variation is driven by         function of markets is weak and that policies
increasing quality and brand differentiation           and regulations strongly influence market
in staples. Wastage in value chains is lower           performance.
than the level found in previous studies (Das
Gupta et al. 2010a, 2010b; Minten, Reardon, and        For instance, Sekhar (2012) finds the extent
Sutradhar 2010; Reardon and Minten 2013).              and degree of market integration is lower for
                                                       rice than for gram or edible oils, and that the
Parastatals dominate food grain marketing,             extent and degree of integration varies by
accounting for over 40 percent of the rice and         region. Sekhar contends that this outcome
wheat marketed surplus. The government’s               results from the substantial regulation and
role goes far beyond its direct involvement            policy intervention in rice compared to
in trade, however. It heavily regulates private        other commodities. Using a wider range of
agricultural trade for staples as well as              commodities and markets, Bathla (2009)
nonstaples. The main rationale of government           finds evidence of only partial integration in
intervention, starting from the 1940s when             wholesale markets across states. While markets
markets were disorganized and very poorly              for some horticultural crops and oilseeds
developed, has been to protect both farmers and        are well integrated, most other markets do
consumers from exploitative or unscrupulous            not demonstrate a consistent, long-term
traders (Shiva 2007; Acharya 2004). Now the            relationship across states. Factors influencing
focus seems to have shifted considerably to            these outcomes include proximity to large
address market distortions created by the same         metropolitan areas, poor infrastructure, and
instruments, effectively segmenting markets            other risks in some states that lead to high
(typically across states), creating disincentives to   price volatility (as in the eastern states). A major
diversify, and prompting calls for action to “get      source of market fragmentation, however, is
the markets right” (GOI 2013b).                        regulations, which remain in place despite
                                                       being relaxed over time. The controls and
Markets within India appear to be relatively           regulations (including movement restrictions
more integrated in the post-reform period              and stock limits) under the ECA, which covers
(Ghosh 2011), yet recent studies of market             all commodities except fruits and vegetables,
integration—which is a reflection of market            are identified as potential impediments to
efficiency—provide a very mixed picture                efficient marketing, including for the major
(Bathla 2008, 2009; Bathla and Srinivaslu 2011;        staples (rice and wheat). The stability imparted
Dasgupta et al. 2011; Acharya et al. 2012; Sekhar      by MSPs, however, clouds the interpretation
2012). For a range of crops and markets, the           of results on market integration, especially
picture that emerges is that internal markets          for states where procurement is concentrated.
are “co-integrated”—that is, they exhibit a            This is consistent with the lack of convergence
long-term relationship. This is not a surprise,        of prices within certain states that are subject
considering that these markets belong to the           to the forces of demand and supply even for
subnational markets within India. A more               cereals (Bathla 2008).
important test is how well they function in
terms of transmitting price signals across space,      The degree of market inefficiency and
and studies consistently find that this key            segmentation, even in rice and wheat markets,



  128
                                              Chapter 12: Current State of Agricultural Markets in India




is highlighted by other recent studies of market      innovations have the potential for overcoming
integration. Analyzing markets at different           some of the well-known constraints to
levels—retail, wholesale, primary wholesale,          marketing efficiency and integrating
and farmgate—Acharya et al. (2012), using             smallholders to value chains. Yet after several
bilateral market pairs, find that price signals are   years of implementation, the scale of these
transmitted at varying speeds across markets.         innovative initiatives remains limited and
The transmission from wholesale to farmgate           their replicability open to question. To better
prices (typically within states) is relatively        understand their performance and draw lessons
quick, but integration across major markets           from their experience, some of the promising
varies widely for both commodities. Despite           initiatives are reviewed here and in the
the influence of policy (MSP) on prices (which is     remainder of this chapter. The specific models
most evident at the farmgate level), empirical        studied include contract farming, cooperatives,
estimates find the highest degree of integration      producer companies, and rural business
between the Chennai and Hyderabad rice                hubs (RBH).
markets. Other markets show significantly less
integration.                                          Contract farming
                                                      In anticipation of improved agricultural
Perhaps the most telling statistic is that
                                                      market performance following the reforms
even in the most integrated market pairing
                                                      and liberalization of the 1990s, the corporate
(rice between Chennai and Hyderabad), only
                                                      sector has shown increased interest in
22 percent of the price divergence from the           processing, exports, and retailing/marketing
long-run equilibrium is adjusted each month,          of agricultural products. A general expectation
suggesting that it would take over 10 months          was that contract farming (CF) would rapidly
for about 90 percent of the price shock to be         expand to better integrate small farmers into
transmitted from one market to another. Most          agricultural value chains, and in general bring
other markets would take longer. Price signals        about much-needed efficiency and growth
thus are not transmitted efficiently across           in the sector. Several corporate initiatives
states, suggesting significant impediments to         in the retail and perishable produce sectors
internal trade.                                       launched CF in the last decade.149 Singh (2012)
                                                      reviews the experience with CF and domestic
                                                      supermarket retail chain linkages, with a focus
Alternative Marketing                                 on smallholders’ participation and on managing
Arrangements: Linking                                 production and market risk.
Smallholders to Markets
                                                      Diversification into high-value agriculture is
Access to markets, especially for small and
                                                      particularly important for the overwhelming
marginal farmers—more than 80 percent
of India’s farmers—is critical to provide
                                                      149	 Foreign investors (such as Metro, Wal-Mart, Tesco, Carrefour)
appropriate incentives for productivity to grow.           have stayed in the wholesale cash and carry business
Despite the isolated pockets of dynamism                   (permitted since 1997) as FDI in retail has been restricted to
                                                           51 percent of the total equity and in single brand retail only
noted here, markets remain relatively weak                 (permitted only since 2010). The recent changes allowing
and inefficient. A number of institutional                 multi-brand retail address some of these issues.




                                                                                                                  129
Republic of India: Accelerating Agricultural Productivity Growth




majority of marginal and small farmers in                           which highlights challenges in enforcing
India, and significant numbers of small farmers                     contracts; undue discounts on value for quality
have indeed diversified into horticultural                          (as perceived by famers); delayed deliveries
crops in irrigated and nonirrigated areas                           and payments; mixed experience with the
(Gaurav and Mishra 2011; Birthal et al.                             provision of extension services, including
2012).150 The main problem facing these                             competitive pricing and the poor quality of
farmers is the inherently higher production                         seed or inputs provided; unequal or unshared
and market risk associated with such crops.                         production risk; an insufficient premium
They lack the economies of scale that could                         relative to open market prices or for quality,
give them an advantage in bargaining power,                         including for organic produce; and a lack
and they face higher transaction costs to                           of attention to social impacts (inadequate
access inputs or market output, as well as                          safeguards against child labor and gender-
basic agricultural services such as credit,                         related issues).
extension, and market information. CF is often
viewed as an appropriate response to these                          The impact of CF practices on natural resource
circumstances, benefiting not only farmers,                         management has also been mixed. In several
but also intermediaries in the value chains and                     instances, CF has promoted more water-
consumers.                                                          intensive cultivation, even in water-stressed
                                                                    areas. For example, the irrigation intensity of
India’s experience with CF has been mixed,                          tomatoes, basmati, and maize grown under
however. Singh (2012) notes that alternative                        CF is greater than that of wheat in Punjab.
modalities have been practiced by different                         Monocultures to ensure year-round supplies
players in India.151 Yields are generally higher                    can dominate CF and deplete soil quality.
among contract farmers (relative to those not                       Insufficient attention to the safe, sustainable
under contract)—but so are costs. Nevertheless,                     use of fertilizers and pesticides has potentially
with generally more stable prices, net income                       adverse impacts on natural resources, the wider
tends to be higher. Birthal and Kumar (2009)                        environment, humans, and animals. Given
document a number of studies showing several                        the short-term nature of CF, issues related
successful CF arrangements (for milk, poultry,                      to sustainability, particularly land and water
spinach, grapes, and gherkins, for example)                         quality, are rarely factored into the business
alongside instances in which CF has been less                       model.
successful or less inclusive of smallholders.
                                                                    Notable exceptions include CF for potatoes in
In practice, the main problems faced in CF                          Gujarat, where McCain Food has promoted the
include default by companies and farmers,                           use of sprinklers without subsidy by all contract
                                                                    farmers. Unilever’s integrated pest management
150	Vegetable crops are the most favored crops on small farms,
     whereas fruits, condiments, and spices are favored on large    package in gherkin production resulted in lower
     farms, owing to the availability of surplus labor, liquidity   fungicide/insecticide use and reduced residues.
     constraints, and a good market price for vegetables.
151	 They include a centralized model (company and farmers);
                                                                    Some CF arrangements for organic basmati rice,
     joint venture with the public sector (NDDB, Markfed); multi-   cotton, and fruits and vegetables have sought to
     partite model (Punjab in many years); nucleus grower model
     (Namdharis); and the intermediate/facilitator model (several
                                                                    reduce environmental pollution and resource
     examples).                                                     depletion (Singh 2012).



  130
                                              Chapter 12: Current State of Agricultural Markets in India




Perhaps the most significant issue is whether         Pro-poor agricultural value chains
CF is inclusive of small farmers. The evidence
                                                      A number of innovations in marketing promote
so far appears mixed. Singh (2012) summarizes
                                                      higher-value crops. To review some of these
studies in India showing that contracting firms
                                                      approaches, especially their ability to link
or entities, mostly private but some public, tend
                                                      smallholders to value chains, Ganguly (2012)
to prefer large and medium farmers for CF. The
average size of contract growers is significantly     analyzes four value chains in Maharashtra:
higher than the average holdings in several           cashew, pomegranate, vegetables, and cotton.
states (see Annex 12, Table A.12.1). Some             The four chains include a producer company
exceptions exist. Firms in Karnataka, Tamil           (for cashew) and three farmer cooperatives.152
Nadu, and Andhra Pradesh work with marginal
and small farmers but tend to focus on specific       Cashew producer company. Promoted by
labor-intensive crops like gherkins. With a           BAIF, a national NGO, a multi-state producer
few exceptions, in the emerging fresh fruit           marketing company in Maharashtra links
and vegetable retail chain it is also medium          farmer cooperatives (largely established by
and large growers who are the “contact” (not          small and marginal tribal households) to
“contract”) growers (Annex 12, Table A.12.2).         organized markets. BAIF’s introduction of
The established or corporate retail chains            cashews among tribal farmers generated modest
prefer the more advanced agricultural states as       improvements in income for poor households,
opposed to the states where the majority of the       but even that small income brought
poor smallholders reside.                             underused land into production and triggered
                                                      socioeconomic change as farmers increasingly
The main reasons for CF to exclude small              ventured into intercropping other cash crops
holdings reflect some of the problems with            like flowers and vegetables. In addition to the
CF mentioned earlier: difficulties in enforcing       gains in income and diversification, many
contracts, challenges in meeting quality              farmers retain some of the raw cashews and
standards (and a high rate of product rejection),     cashew apples for their own consumption,
high transaction costs, and smallholders’             which adds to nutritional security. The initiative
weak bargaining power. Consistent with                has also generated downstream employment
these findings, Birthal and Kumar (2009) note         in primary processing facilities, benefiting
that successful CF cases involved market-             both the member households and landless
linked prices to avoid opportunistic behavior,        households.
providing essential services like insurance and
banking (interest-free advances, capital for start-   Pomegranate cooperative. The producer-
up costs, and so on), operating through farmer        led Krishi Vikas Cooperative (KVC) was
groups or intermediate contracts to overcome          registered in 1989 and later federated with five
the high costs of transacting with individual
farmers producing small volumes, and nonprice         152	 The case studies included in the study are cashew—
factors (such as the regularity of off-take and            Vasundhara Producer Company Limited (VAPCOL) in Pune,
                                                           promoted by BAIF, a national NGO; pomegranate—Krishi
payments, on time delivery of inputs, and                  Vikas Cooperative, Sangola, Solapur, promoted by farmers;
technical advice). These relationships are                 vegetables—Nandini Vegetable Producer Cooperative,
                                                           Nandni, Kolhapur, promoted by farmers; and cotton—Indira
established to exploit long-term mutual benefits           Gandhi Sut Girini Cooperative, Wardha, promoted by
by both the agri-business and the farmer.                  farmers.




                                                                                                              131
Republic of India: Accelerating Agricultural Productivity Growth




other cooperatives to form Maha Anar. The            Cotton cooperative. Indira Cooperative Sut
cooperative successfully created backward and        Girni Limited at Wardha was one of many
forward linkages for its member growers but          yarn-manufacturing units promoted by the
failed to retain the trust of its members and        Government of Maharashtra under the Fifth Five
expand business. Membership declined, and            Year Plan to integrate producers in value chains.
members started selling their produce through        The major objectives were to purchase member’s
other channels. Focus group discussions              produce, provide extension services, and
revealed that small farmers participated little      establish research and development activities.
in pomegranate cultivation throughout the            Ginning and pressing activities started in 1985.
area owing to irrigation problems and the            With a huge capital investment in ginning
riskiness of the crop. When small and marginal       and yarn-making and with very low operating
farmers take up pomegranate cultivation,             margins, the business became unsustainable
however, their incomes improve significantly.        and prone to high market risks. To become
KVC is a partial success as a pro-poor value         sustainable, the business requires professional
chain. The cooperative provided initial access       management along the components of the
to domestic and international markets for            value chain and diversification into higher-value
farmers in the area and expanded cash cropping       products such as dyed yarns and textiles.
among smallholders. As for many cooperatives,
the challenge for KVC appears to be weak             Lessons. A number of lessons emerge from
                                                     these experiences. In all of the value chains,
governance. The priority now is to regain
                                                     at least initially, small farmers benefited from
members’ confidence.
                                                     gaining access to markets, information on
                                                     better production techniques and markets, and
Vegetable cooperative. The Nandini
                                                     higher returns. The value chains introduced
cooperative in Kolhapur was formed to organize
                                                     new crops, and farmers overcame barriers to
small-scale vegetable growers and link them
                                                     entry such as the need for irrigation and other
to national markets. The cooperative started
                                                     investments to diversify. Crop diversification
procuring vegetables directly from farmers in
                                                     is an important foundation for building pro-
1986. By 1993, instead of having to transport
                                                     poor value chains—farmers will take risks for
their produce by truck, farmers benefited from
                                                     profitable and sustainable activities. Farmer
collection centers developed by the cooperative      initiative also matters. For example, cashew
(1 center for every 10 villages). The cooperative    farmers intercropped to generate additional
successfully linked vegetable growers to             income, whereas pomegranate farmers did
various national markets, especially in the          not. All four value chains, however, failed to
first two decades of its existence. It created       introduce or adapt appropriate technology
the required infrastructure and demonstrated         over time. Finally, producer companies seem to
a proven mechanism to market vegetables in           offer a better organizational framework than
distant markets. Over time it failed to sustain      the cooperative model for linking small-scale
the system, however, and its business declined       producers to markets.153
drastically. The lack of skills and capacity to
manage the business under a viable business          153	 For example, the producer company (VAPCOL) fared better
                                                          compared to the cooperative (KVC) venture as far as forward
plan is a critical constraint on the cooperative’s        linkages are concerned. VAPCOL engaged professional
ability to sustain its growth.                            managers for marketing activities.




  132
                                                           Chapter 12: Current State of Agricultural Markets in India




Efficacy of Market Channels in                                     and the most popular crop there—wheat—is
Improving Access                                                   highlighted here. Farmers have several options
                                                                   to purchase seed: traditional input dealers,
The structure and functioning of markets at                        PACS (primary agricultural credit society, a
the village level is important for promoting                       state organization of farmers), state seed stores
technology and productivity. A survey-based                        (located mainly at the district level), agricultural
study of three alternative market channels—                        universities (offering direct retail of breeder
traditional private sector traders, state-sponsored                and foundation seed to selected farmers),
cooperatives, and Rural Business Hubs (RBHs)—a                     RBH outlets, and mandi (wholesale market)
modern private sector innovation)—by Reardon                       traders. The major share (over 80 percent) of
et al. (2012) provides a descriptive and statistical               the seed market is supplied by the private
analysis of the performance of these channels.                     sector (56 percent through traditional retail
The study specifically investigates the outreach                   and 24 percent by RBHs). Marginal farmers who
and incidence of services and proximate impacts
                                                                   bought wheat seed in Uttar Pradesh purchased
on technology use and productivity at the farm
                                                                   12 percent of it from state/cooperative
level. The study involved three states (Andhra
                                                                   stores and 4 percent from other farmers; the
Pradesh, Madhya Pradesh, and Uttar Pradesh) to
                                                                   overwhelming majority bought it through
capture different agro-ecological, economic, and
                                                                   private channels (21 percent from HKBs and
institutional environments.154 The RBHs included
                                                                   63 percent from small traditional shops).
are Hariyali Kisaan Bazar (HKB) in Uttar Pradesh,
                                                                   Clearly a minority of poor and marginal farmers
ITC’s Choupal Saagar in Madhya Pradesh, and
                                                                   purchase seed from state/cooperative outlets.
Viswas in Andhra Pradesh. These firms entered
the scene around the mid-2000s. The major
                                                                   Viewed from a sales perspective, the findings
findings and conclusions concerned seed and
                                                                   are similar. State/cooperative outlets sold 16
input markets, input and output prices for small
                                                                   percent of wheat seed to marginal farmers and
and marginal farmers, and the determinants
                                                                   22 percent to small farmers; 62 percent of sales
of purchases and prices in different market
channels.                                                          were to medium farmers. This finding sharply
                                                                   contradicts the conventional view that the
                                                                   (subsidized) state/cooperative outlets focus on
Seed, fertilizer, and farm chemical                                selling to the poor. Although to some extent
markets: Subsidized outlets sell                                   this sales pattern reflects the land distribution,
subsidized inputs mostly to medium-                                which is dominated by medium farmers, the
scale farmers                                                      targeting of the seed subsidy does not appear to
The findings for seed markets are similar across                   be pro-poor.
all three states, so the case of Uttar Pradesh
                                                                   Similar conclusions emerge for fertilizer and
154	 The study chose six zones by reasoned sampling around
     nodes (RBH centers). The survey sample is therefore not
                                                                   farm chemical markets, with a few exceptions.
     random and may not represent markets more generally (for      In Uttar Pradesh, the major share of sales from
     example, markets where RBHs are not located). Nevertheless    state/cooperatives of the heavily subsidized
     a comparison of the three channels provides important
     and interesting insights. For each state, household surveys   fertilizers (73 percent) and chemicals (83
     were conducted in 30 villages within and just beyond          percent) are to medium farmers (73 percent).
     the catchment areas of six RBHs equally distributed over
     western, central, and eastern study zones in each state.      One exception in Uttar Pradesh is RBH sales



                                                                                                                 133
Republic of India: Accelerating Agricultural Productivity Growth




of urea, two-thirds (67 percent) of which are       the case of poorer Eastern Uttar Pradesh, where
to marginal/small farmers. In general, RBH          the probability of using state sources increases.
sales follow the pattern of land distribution,      For other parts of Uttar Pradesh, as well as in
with their main client (by volume share) being      Madhya Pradesh and Andhra Pradesh, however,
medium and large farmers.                           the likelihood of purchase from a state/
                                                    cooperative outlet falls sharply for marginal
Input and output prices do not place                farmers but rises with education, being a
                                                    cooperative member, having wealth through
small and marginal farmers at a
                                                    livestock, and the distance to traditional stores.
disadvantage
Input prices are consistent across farm-size        Two other notable results are the greater
groups, so marginal and small farmers do not        likelihood that minorities (“non-Hindus”) will
pay more than larger farmers for inputs—            purchase fertilizers from state/cooperative
another finding that is contrary to expectations.   stores, suggesting that they tend to be more
Similarly, output prices show no systematic         socially inclusive, perhaps reflecting the
variation across farm size groups. The mandi        location effects of alternative channels. For
wholesaler, RBHs, and brokers paid close to         pesticides, the probability of purchase from a
the same price for wheat and soybeans (net          modern outlet (RBH) is positively correlated
of transaction costs) irrespective of farm size.    with farm size, but there is a distinct U-shaped
Although small and marginal farmers are             probability curve for herbicides, with both
presumed to dominate rural food markets,            marginal and medium farmers more likely to
such farmers account for only a small volume        buy from RBHs.
of output handled by those markets. Mandis
and RBHs thus rely little on output supplied by     Price formation. In Uttar Pradesh, paddy seed
small and marginal farmers.                         prices paid by farmers decline with greater
                                                    education, the age of the household head,
Determinants of purchases and prices                nonfarm income, farm machinery, greater
                                                    distance from the district town, and being in
in different market channels
                                                    the favorable western and central areas. These
Choice of market channels for inputs. The           factors may suggest greater bargaining power
choice to purchase inputs from a particular         among farmers, but they could also reflect the
market channel is influenced by proximity           size and level of development of the market.
(distance to the respective outlet for the market   By contrast, worse prices are paid by scheduled
channel) as well as nonfarm income (off-farm        castes and tribes and by farmers in the eastern
employment). For example, the likelihood of         zone. In Madhya Pradesh, there is a strong
a farmer choosing an RBH outlet for paddy           negative association of RBH with fertilizer and
seed in Uttar Pradesh increases with off-farm       pesticide prices, again likely reflecting the
employment, distance to traditional stores, and     effects of modernization and larger transaction
proximity of the RBH. But marginal farmers are      volumes.
significantly less likely to be RBH customers,
suggesting an “asset bias” in the modern            In general, farmers in eastern zones do not
market. Again, there are some exceptions, as in     benefit from modern market channels as much



 134
                                              Chapter 12: Current State of Agricultural Markets in India




as the other groups studied. For example,             ECA, and the requirement that all produce
farmers in the poorer eastern zone pay more           be sold at a limited number of licensed and
for their pesticides (as in Uttar Pradesh), and the   regulated markets under the purview of the
likelihood of output purchases by RBH falls with      committees set up under the APM(D&R) Act,
being located in the east or belonging to SC/ST       referred to as APM Committees (APMCs), have
social classes. The relatively better-off farmers     created significant disincentives for private
(in terms of assets and livestock) are better         investment in storage, handling, and marketing
connected and more likely to participate in           infrastructure. They have also promoted the
modern output markets, while “poorer” farmers         fragmentation of markets (by restricting
rely on selling primarily to the traditional          movement and through cess/tax implications)
sector.                                               and made market transactions and price
                                                      discovery less transparent.
Implied quality of inputs. Since quality cannot
be determined from sales data, an indirect            These regulations constrain direct marketing
assessment is to infer quality using partial          arrangements such as CF or direct purchases
factor productivity measures. The results             by food-processers from farmers, and they
do not suggest any obvious discrimination             prevent improvements in value chain
against smallholders in terms of quality. For         efficiencies. Thus the very considerable
example, for paddy and wheat in Andhra                potential for food processing remains largely
Pradesh, seed purchased from the traditional          untapped, dominated by low-level processing
market is associated with higher output than          by informal enterprises (Bhavani, Gulati, and
the same seed purchased through state market          Roy 2006; Morisset and Kumar 2008a, 2008b).
channels. In Madhya Pradesh, farmers buying           Farmers as well as consumers bear the brunt
inputs within traditional markets show lower          of these regulations, which have fostered
but comparable returns to increases in farm           uncompetitive market structures at the mandi
operations, which likely reflect other attributes,    level, depressing farmer prices and raising
but no obvious association suggests comparable        final consumer prices.
quality across farm-size groups. In Andhra
Pradesh, farmers’ main problem appears to             The issues related to the ECA and APMCs are
be a labor shortage rather than any problems          well recognized and widely acknowledged. To
directly associated with other inputs.                address the major problems, the government
                                                      introduced a “Model Act” in 2003 and urged
                                                      state governments to amend their respective
Status of Agricultural Marketing                      legislation and regulations accordingly.155
Reforms                                               Most states have amended their legislation
                                                      to reflect the recommendations of the Model
Among the controls that have “stifled”
                                                      Act, but the nature and extent of reforms (and
agriculture (GOI 2013b), two pervasive
                                                      more importantly, their implementation) by
regulations are the Agricultural Produce
Markets (Development and Regulation) Act—
APM (D&R) Act—and the Essential Commodities           155	 The APMC is a “State Subject,” so the central government’s
                                                           role in implementation legislation and regulations relating to
Act (ECA) (see Annex 12 for a brief description).          the APMC (as with other agricultural regulations) is limited,
The zoning and storage restrictions under the              with responsibility resting squarely with the states.




                                                                                                                  135
Republic of India: Accelerating Agricultural Productivity Growth




individual states is not known. Similarly, the     Despite the central government’s 1993
government has periodically modified the           decision to treat the entire country as a single
ECA provisions to ostensibly pare down the         food zone, markets remain segmented. The
list of “essential commodities.” Again, with       issues that continue to constrain markets
limited central government control over            and trade, by raising transaction costs and
implementation and authority for each state        creating disincentives for private investment
to establish its own regulations and controls,     in marketing and trade infrastructure, are
progress on removing the ECA’s restrictions on     many. Even though ECA currently has no
domestic markets and trade is not known.           movement and stock restrictions, several states
                                                   have established ad-hoc stock restrictions,
To address this knowledge gap, Patnaik             with frequent changes, creating uncertainty
and Sharma (2013) conducted a study that           and raising transaction costs. A lack of
also included detailed field assessments in        transparency prevails: Orders can be issued
five states. The main findings are not very        independently by state governments and
encouraging. Although a few states moved well      sometimes by different departments within the
ahead in implementing both the letter and          same state. Because no compilation exists of
spirit of the APMC reforms, implementation is      all orders in effect at any time, it is extremely
generally uneven and inconsistent across states.   difficult to get a sense of what is happening
The net result is that despite some progress,      on the ground. Despite their best efforts,
significant impediments still constrain the        Patnaik and Sharma (2013) could not obtain
performance of markets and internal trade. The     or compile a comprehensive list of all orders
major findings from the study are reported in      currently in effect.156 Multiple and ad hoc taxes
the sections that follow.                          are another problem. Several statutory, fiscal,
                                                   and administrative barriers hamper the free
Current provisions of the ECA. The list of         movement of agricultural products between
essential commodities has changed over the         and within states. They include border check
years, with the latest list, issued in 2006,       gates (entailing avoidable delays that are costly
pruned down (in principle) to only seven           for products that are perishable or have a
items (Box A.12.1). Qualitatively, the coverage    short shelf-life); “entry” and “export” taxes in
of agricultural commodities has changed            some states; other state-specific taxes; multiple
little, as the key provision of the broadly        taxation (taxes levied by different markets
defined “foodstuffs” essentially covers all        within each state); and a lack uniformity in
food commodities, whether processed or             fiscal measures across states.157
unprocessed. The act also covers key inputs,
fertilizers, and seeds for most major crops        156	 As statutes and regulations in India generally do not have a
(certainly all food crops). Equally important,          “sunset clause” and continue to be in force until their repeal,
                                                        it is likely that most of these orders remain in force.
the list of commodities can be amended by          157	 For example, in Punjab the total market charges on food
executive orders from the central and state             grain transactions are around 15.50 percent (a market fee of
                                                        2 percent, development cess of 2 percent, purchase tax of
governments. The result is a multiplicity of            4 percent, commission charge of 2 percent, infrastructure
Control Orders and frequent changes of rules            cost of 1.5 percent, and VAT of 4 percent) ad valorem, apart
                                                        from the charges for weighing-(Rs. 0.55), loading (Rs. 0.40),
and restrictions (see Patnaik and Sharma 2013           brokerage (Rs. 0.16), hamali (labor in the market area) (Rs. 1),
for recent examples).                                   and cleaning (Rs. 0.65 per bag per quintal).




 136
                                               Chapter 12: Current State of Agricultural Markets in India




APM (D&R) Act reforms as per the Model Act            also adopted other provisions, such as bulk
(2003). Of the 35 states and union territories,       purchases (Uttar Pradesh) and contract farming
6 had never adopted the Act and required no           (Punjab), but these are on an exceptional basis
reforms (see Annex 12, Table A.12.3). Of the          through executive orders. On the other hand,
remaining states and territories, 18 formally         Maharashtra is the most progressive state in
amended their Acts, and 1 (Bihar) completely          terms of market reforms, and it has amended
repealed it in 2006. The remaining 11, including      the APM Act but does not provide for single
some major agricultural states (Punjab, Uttar         point levy of the market fee due to opposition
Pradesh, West Bengal) have not adopted the            by the APMCs.
Amendment. The Model Act contains five
critical provisions for liberalizing trade:           Special provisions (such as market fee
                                                      exemptions in Punjab) exist for particular
   1.	 Direct marketing (purchase by the              businesses, often business that have promised
       buyers outside market yards).                  to make large investments in the state. Case-
   2.	 Contract farming.                              by-case exemptions through executive orders
                                                      may be expedient compared to enshrining the
   3.	 Entry of private sector (to establish          provisions in the statute itself, but they do
       wholesale markets).                            not allow robust market development. Such
   4.	 Single license (for operating in more          exemptions are surrounded by uncertainty
       than one APMC jurisdiction).                   because the orders can be rescinded at any
                                                      time—often for political reasons. For example,
   5.	 Single levy of market fee (irrespective of     the Government of Punjab allowed Cargill to
       the times produce is sold and purchased).      make direct purchases of food grains from the
                                                      mandis, but the order was withdrawn within
On paper, progress toward the much-anticipated        a year following opposition by commission
market reforms appears to be good—most                agents.
states have amended the APM(D&R) Act—but
real progress remains much more modest.               Finally, for actual implementation, the
First, the states have not adopted all provisions     amendments to the Act need to be formulated
of the Model Act (Annex 12, Table A.12.3).            as rules and notified. In this regard, progress
Only 6 of the 18 amending states adopted              is even weaker. Annex 12, Table A.12.4
all 5 key provisions, while others adopted            shows that only 10 of the 18 adopting states
the amendments only partially. The uneven             have formulated and notified the rules for
implementation of specific provisions gives           implementation. Again, the actual provisions
mixed and unclear picture.                            and rules vary across states. Maharashtra and
                                                      Karnataka have detailed rules in place for
Some states have not formally adopted                 implementation. In Andhra Pradesh, rules
the amendments but have made specific                 for direct marketing have not been covered
provisions through executive orders on a              in the notification. Because Madhya Pradesh
case-by-case basis. For example, both Uttar           has exempted sales of fruits and vegetables
Pradesh and Punjab have provisions for a              (except bananas) from the provisions of the
single-point levy of the market fee. Both have        Act, it needs no rules, since the provisions of



                                                                                                   137
Republic of India: Accelerating Agricultural Productivity Growth




the Act are not now applicable to fruits and             significant private sector response in terms of
vegetables. Himachal Pradesh has no restriction          initiatives involving contract farming, direct
for setting up private markets; licenses for             purchase, and the establishment of mega-food
private markets and direct marketing are issued          parks (Patnaik and Sharma 2013). Some of the
on a case-by-case basis. Rules framed by Odisha          private sector initiatives in different states are
for private markets specify no restrictions on           summarized in Annex 12, Table A.12.5. Note
setting up private markets, except to state              that investment and corporate activities have
that the location for the private market will            been more common in states where reforms
be specified by the state government and will            were more significant.
not be permitted within one kilometer from
an existing APMC. Such restrictions are not              Among the states, Maharashtra has been
conducive to the spirit of the reforms, which is         the front-runner in market reforms and has
to increase effective competition.                       seen significant new initiatives and impacts
                                                         (Box 5), yet even in Maharashtra challenges
Impact of reforms. As indicated, whenever the            remain in implementing the provisions of the
opportunity has arisen—either through formal             amendments. For instance, despite several
reforms or through special provisions—the                government initiatives, contract farming
private sector’s response has been swift and             remains informal. Maharashtra is at the
at times revolutionary. The improvements in              forefront in establishing private markets but
potato and rice supply chains are examples               faces a number of problems. On the one hand,
(Reardon and Minten 2013). Reforms have                  farmers do not perceive the benefits of selling
spurred other innovations such as the                    their fruits and vegetables to the dedicated
emergence of virtual markets (including futures          private market set up for them, as there are no
markets, spot exchanges, warehouse receipt               additional facilities or services. On the other
systems, ICT-based market information, and               hand, the private market has run into several
web marketing), electronic spot exchanges, and           problems with the local APMC, restraining its


Box 5: Progress in market reforms in Maharashtra

 The Maharashtra State Agricultural Marketing Board actively promotes public-private partnerships in
   setting up markets and infrastructure.
 A total of 23 private markets are functional, of which the first private market for fruits and vegetables
   has been set up by Premium Farm Fresh.
 Maharashtra issued 47 licenses for direct marketing to companies/traders.
 Licensed E-markets under APMC Act, Maharashtra have led to the establishment of spot exchanges by
   MCX, NCDEX, and Reliance Spot.
 Trade in fruits and vegetables has been deregulated and exempted from market fees.
 Paithan Mega Food Park has been given special status for sourcing raw materials.
 A total of 10 contract farming sponsors have been registered.
 Major private companies are involved in contract farming, such as Marico (Saffola), ITC and Pepsico
   (potatoes), More, Reliance, Big Bazaar (fruits and vegetables), and KB Export (vegetables for export).
Source: Patnaik and Sharma 2013.




 138
                                                    Chapter 12: Current State of Agricultural Markets in India




ability to provide comparable services. APMC                providing good services, infrastructure, and
contends that any produce in its jurisdiction,              public utilities.
even if traded in the private market, needs to
pay APMC market fees. APMCs have opposed the                More tangible benefits of reforms are
implementation of a single-point levy for market            demonstrated by the specific impacts on the
fees, and the private market is not allowed                 benefits to producers as well as consumers
by the APMC to procure produce at collection                from better-functioning markets, indicating the
centers. All of these stipulations reduce the               large potential impact such reforms can have.
revenues or raise costs to the private operator,            Table 23 presents an example of the benefits of
restricting the viability of the business.                  competition gained by allowing private mandis
                                                            to operate. Competition has driven down the
In contrast, Tamil Nadu has not amended its                 market and other fees charged for the same
Act but already had a provision for setting                 transactions, as APMC is now forced to compete
up private markets and for direct marketing.                for customers. Another specific example is the
The Agricultural Products Producers and                     reduction in marketing margins through direct
Traders Association Market, a modern fruit                  marketing, which benefits farmers (receiving a
and vegetable market, is performing well (in                higher output price) and consumers (paying a
strong contrast to Maharashtra’s experience),               lower retail price), shown in Table 24.

Table 23: Comparison of charges in private and APMC markets
                                                                                 Private mandi (Santosh Pvt.Ltd
Charge                                           APMC Pimpalgaon                         Bagulgaon Yeola)
Market fees/charges                     1%                                     0.75 %
Supervision charges to GOM              0.05%                                  0.05%
Commission charges                      4%                                     4%
Weighing                                Rs. 1 per 100 kg                       Rs. 1 per 100 kg
Hamali                                  Rs. 4 per 100 kg                       Rs. 2.50 per 100 kg
Sorting/grading                         Rs. 20 per 100 kg                      Rs. 6 per 100 kg
Loading charges                         Rs. 4 per 100 kg                       Rs. 7 per 100 kg
Source: Patnaik and Sharma 2013.
Note: GOM = Government of Maharashtra.


Table 24: Price build-up in tomato value chain (Salarpur, Uttar Pradesh)
                                        Traditional marketing                  Direct marketing (Vista Foods)
Particulars                            Rs/q                 %                     Rs/q                  %
Producer share                       800.00              43.24                  900.00               53.42
Market fee                            11.50                0.62                   9.00                0.53
Commission charge                    102.36                5.53                   0.00                0.00
Logistic and OH cost                 290.00              15.68                  387.00               22.97
Retailer/processor margin            646.14              34.93                   388.8               23.08
Consumers price                     1850.00             100.00                 1684.80              100.00
Source:  Patnaik and Sharma 2013, based on key informant interviews in UP case study.




                                                                                                            139
Republic of India: Accelerating Agricultural Productivity Growth




Conclusions and Implications                       are to be shared will be an area for policy
                                                   attention.
In most cases, market reforms have not been
implemented in full or in earnest and are          Overregulation is clearly demonstrated to
instead hampered by provisions and omissions       restrict private investment and trade, as seen
that have effectively retained the status quo.     from the experiences of Maharashtra compared
The APMCs continue to exert strong control         to Uttar Pradesh and Punjab. Even initiatives
and have been a key obstacle to reform,            to promote private investments, such as the
even where states have amended their               schemes for “Modern Terminal Markets” and
respective Acts.                                   “Wholesale Markets,” have not been successful.
                                                   The subsidies offered in these instances are
A key message emerging from the analysis is        clearly not attractive enough to overcome the
that India remains a segmented agricultural        unattractive restrictions these schemes place on
market, with high transaction costs that hurt      private sector activities.
both producers and consumers, hinder the
smooth internal flow of trade, and restrict        An important problem is the conflict of interest
productivity growth. Removal of movement,          for state governments. The current markets
stock, and trade restrictions for smoother         and APMCs generate revenue. Any change
internal trade remains a very high priority.       in market regulations will have resource
                                                   implications for state governments, which
The experience with different states               must be dealt with in terms of sharing the
interpreting and implementing reforms and          costs of reforms. Another important conflict of
regulations in different ways suggests a need      interest is the state departments or marketing
to rethink market governance. For example,         boards responsible for both running and
differential and multiple taxation, different      regulating the markets. It is important to
and frequently changing stock limits, and          ensure an independent market regulator to
movement restrictions create problems in           create a level playing field between the APMCs
marketing across state borders. Agriculture        and private markets.
is a State Subject, but Inter-State Trade and
Commerce is under the Union List. It may           Finally, while the overregulation inherent in the
be advisable to consider putting agricultural      current APM(D&R) Act is clearly a hindrance,
marketing on the Concurrent List and               the absence of all market regulation would
establishing common norms for taxation             not be desirable. An important role of the
and other charges/fees to make the system          government is oversight and regulation, and
more transparent and predictable. This             the complete abdication of this role may not
approach will facilitate smoother movement         lead to a better outcome. For example, since
of agricultural products across states, foster     Bihar completely repealed the APM(D&R) Act,
more efficient modes of marketing through          its markets have been completely unregulated.
electronic trading and virtual markets, and        Bihar sought to promote free, competitive
help achieve the objectives of the Model Act.      transactions, but the state’s public facilities
Reform of fiscal policies will obviously have      are public goods; if no entity is responsible
revenue implications, and how these costs          for them, there will be no investment in



 140
                                            Chapter 12: Current State of Agricultural Markets in India




infrastructure or maintenance to improve           given reasonable assurances that such
market conditions. Most importantly,               investments would be secure and generate
governance of the markets has become a critical    returns. An independent market regulator is
issue, as the markets allegedly have been taken    clearly needed, with the primary objective
over by a few traders and cartels.                 of promoting competition and creating an
                                                   orderly environment for price discovery and
To create a healthy environment for                physical transactions, and with the capacity to
transactions, it is important for markets to       provide basic services as needed (perhaps until
be orderly and governed well. The private          the private sector steps in), including dispute
sector would invest in market infrastructure,      resolution services.




                                                                                                141
13                Agro-Industry: The Food-
                  processing Sector




A
        s the structural transformation of          backward multipliers (both output and income
        the economy accelerates in India, the       multipliers) based on the national Input-Output
        nature of agriculture and the food          tables constructed by CSO (Ganesh-Kumar
industry will change dramatically. With             and Panda 2013). The employment multiplier
inevitable urbanization, the shift to nonfarm       effects from investments in the food-processing
activities, rising incomes, women’s increasing      industry are also high—about 2.5 times that of
participation in the labor force, and changing      other industries (Bhavani, Gulati, and Roy 2006).
consumption habits, an outpouring of demand         Finally, processed exports are an important
for processed (including prepared and packaged)     outlet for agricultural output. Exports of
foods will occur. For India, food processing        processed foods now constitute almost
provides a natural entry point for the sluggish     48 percent of rapidly expanding agricultural
“manufacturing” sector to enter agricultural        exports, almost doubling in (real) value over the
areas, exploit their present comparative            past three years (Figure 73).
advantage, and create much-needed off-farm
employment.                                         To capture these benefits, the central
                                                    government has made concerted efforts since
A well-developed food processing industry can       the mid-1990s to attract private investment in
also stimulate higher agricultural productivity     the food-processing industry by giving it priority
and growth through several channels. Examples       sector status, providing fiscal incentives,158
include better and more stable farm prices;         and initiating other reform measures. The
reduced wastage as produce unsuitable for wet       proposed National Mission on Food Processing
markets is processed into more value-added          will scale up these efforts at the state level, with
consumables, increasing returns to farmers; and     technological and logistical support from the
diversification into crops needed for processing,   center. The food-processing sector has grown
potentially transforming traditional “food”         rapidly in the past five years, with investments
crops into “cash” enterprises.                      growing at 20 percent per year. Foreign direct

The importance of food processing for               158	 Some of these include a five-year tax holiday and 35 percent
                                                         tax deduction for the same period for setting up new agro-
stimulating growth and employment stems                  processing industries; reduced import duty on processing
from its high backward linkages. Of all                  machinery; no corporate taxes on profits from export sales;
                                                         and automatic approval for 100 percent FDI in most items.
activities in the economy-wide activity matrix,          Exemptions from excise duty have been given to encourage
food processing has the highest estimated                capital investment in large projects and processing firms.




                                                                                                               143
Republic of India: Accelerating Agricultural Productivity Growth




Figure 73: Composition of food exports, 2012/13                       Food Industry Structure and
                 Floriculture, 0.6%   Fresh F&V, 4.8%                 Investment
                                            Processed F&V, 4.2%
                                                                      Like all manufacturing in India, the food-
                                                                      processing sector has a dualistic structure: A
                                                   Animal Products,
Cereals, 44.0%                                     16.7%
                                                                      relatively small (in number of units) but capital-
                                                                      intensive organized segment coexists with a
                                                                      pervasive, mostly rural, and relatively labor-
                                                                      intensive unorganized segment. Organized
                                                                      and unorganized food-processing operations
                                                Other Processed       employed nearly 8 million people in 2008–09
                                                Food, 29.6%
                                                                      (the latest year for which survey data are
Source: Authors, using APEDA data.                                    available)162 and contributed over Rs. 500 billion
                                                                      (about US$ 10 billion) to the economy (valued
investment (FDI) is still relatively small (US$ 2                     added in 2004/05 prices).163 In 2008/09, the
billion cumulative investment between April                           organized segment provided 1.56 million jobs in
2000 and August 2013) compared to the overall                         a little over 27,000 units, producing an average
investment in food processing of about US$ 24                         of Rs. 13 million of output per unit (in real
billion, but it is expected to grow more rapidly                      2004/05 rupees). In the unorganized segment,
with the expansion of organized food retail.159                       about 2.6 million enterprises employing
                                                                      6.3 million people producing Rs. 0.054 million
The literature on agro- and food processing in                        of output per enterprise in 2005/06.
India is limited and mostly outdated. Given the
importance of food processing in promoting                            Figure 74 shows the dualistic structure of
agricultural productivity and generating                              the food-processing sector in 2005/06. The
employment, this chapter analyzes the food-                           unorganized segment dominates in numbers
processing sector—its structure, evolution,                           of (small) enterprises and workers, but the
and the interplay of factors and policies that                        organized segment dominates in terms of the
influence its performance.160 To place India’s                        value of output and investment. The average
food-processing industry in the context of other                      scale of operations is much smaller in the
industries, the analysis will rely on indicators                      162	 The analysis of organized manufacturing is based on state-
such as the number of units (factories),                                   level data from the Annual Survey of Industries (ASI) from
                                                                           1980–81 to 2008–09. Data used are for the food and beverage
employment, investment, capital stock, and                                 industry (National Industrial Classification code 15) for
gross value added (GVA).161                                                17 major states of India. The analysis of the unorganized
                                                                           segment uses the 2000–01 and 2005–06 quinquennial NSS
                                                                           data (the 56th and 62nd rounds). The analysis for organized
159	 FDI estimates from the website of the Department of                   food manufacturing is confined to 17 major states and that
     Industrial Policy and Promotion, Ministry of Commerce and             for unorganized manufacturing to 20 major states at 2004–05
     Industry, GOI. Growth estimates from India Brand Equity               prices. Of 24 groups of industries at the two-digit level as per
     Foundation.                                                           the National Industrial Classification (NIC) 2004, unit-level
160	 This chapter is based on a background paper by Bathla and             data on food products and beverages (NIC 15) are extracted
     Gautam (2013).                                                        for selected states. For additional detail on the data and
161	 All values are converted to real terms with the base year             variables used in the analysis, see the background paper.
     2004–05. Investment is taken to be the difference in fixed       163	 The size of the organized segment was Rs. 352.5 billion
     capital in two periods plus depreciation in year t. Capital           in 2008/09, and the size of unorganized segment was
     stock is estimated using the Perpetual Inventory Method.              Rs. 142.5 billion in 2004/05 real rupees.




  144
                                                                Chapter 13: Agro-Industry: The Food-processing Sector




            ualistic structure of the agro-
Figure 74: D                                                                  Share of agro-processing in organized
                                                                   Figure 75: 
           processing sector                                                  manufacturing
                    Structure of Agro-Processing Sector            25

      GVA                                                          20

                                                                   15
Employment
                                                                   10

                                                                    5
      Units
                                                                    0




                                                                                                                          Investment
                                                                          Factories




                                                                                        Employment




                                                                                                      Gross Value Added




                                                                                                                                       Capital
            0%        20%      40%       60%       80%     100%
                                                                    -5
                 Unorganized (2005/06)    Organized (2005/06)
                                                                   -10
Source: Bathla and Gautam 2013.
                                                                   -15

unorganized segment, with fixed assets per                         -20
enterprise of Rs. 0.096 million employing an                                          1980-89        1990-99              2000-09
average of 2.4 persons, compared to Rs. 25.6                       Source: Bathla and Gautam 2013.
million and 57.4 persons per enterprise in the
organized segment.
                                                                   of employment in organized manufacturing
The structure of unorganized manufacturing is                      on average between 2000 and 2008, with a
changing among the three types of enterprises                      steady share in output of 8–9 percent over time
defined by NSS: Own Account Manufacturing                          (Figure 75). The shares in investment and capital
Enterprises (OAMEs), Non-Directory                                 stock are lower but grew from 5.5 percent
Manufacturing Establishments (NDMEs), and                          in the 1980s to 8 percent in the 2000s. The
Directory Manufacturing Establishments (DMEs)                      relative size of food processing varies by state
(defined in Annex 13). Small, family-run OAMEs                     (Annex 13, Table A.13.1a), with few changes
dominate the unorganized segment (81 percent                       over time. The share of food processing in
of enterprises) and are located primarily in                       investment and output in lagging states (Bihar,
rural areas (the share of NDMEs is 13 percent                      Madhya Pradesh, and Rajasthan) as well as some
and that of DMEs 6 percent). OAMEs also had                        advanced states (Punjab and Andhra Pradesh)
the largest share of output (42 percent) in                        has increased. Most states show no discernible
2005–06, compared to NDMEs and DMEs (25 and                        trends, however.
33 percent). Over time, OAMEs have gradually
declined, and the share of DMEs in almost all                      In the unorganized segment, food processing
the key indicators has expanded.                                   has a slightly higher share of the industry
                                                                   (17 percent of employment and value added in
                                                                   2005/06). Of the 82,897 unorganized enterprises
Spatial and temporal trends
                                                                   surveyed in 2005/06, a little over 15 percent
Food processing provides significant                               of all enterprises (rural and urban) were in
employment in both the organized and                               food processing, accounting for 18 percent
unorganized segments. It accounted for                             of investment in the sector (Figure 76). Rural
about 18 percent of factories and 16 percent                       unorganized food-processing accounts for



                                                                                                                                        145
Republic of India: Accelerating Agricultural Productivity Growth




           Share of agro-processing in
Figure 76:                                                                 evident in Table A.13.2. While the shares
           unorganized manufacturing                                        of employment and numbers of units and
35.00                                                                       enterprises have been relatively stable over
                                                                            time across states, some states have become
30.00
                                                                            more productive, with growing shares in GVA
25.00                                                                       for food processing—notably Andhra Pradesh,
20.00                                                                       Karnataka, Punjab, and Rajasthan.
15.00
                                                                            The differences in shares across states,
10.00
                                                                            especially for employment, may simply
 5.00                                                                       reflect differences in the size of each state’s
 0.00
                                                                            population. Figure 77 shows that with the
            Rural Urban Rural Urban Rural Urban Rural Urban                 exception of a few “outliers” (circled in the
             Enterprises  Workers    Investment     GVA
                                                                            figure), the more populous states are also
                             2000/01    2005/06
                                                                            generally poorer. Organized employment is
Source: Bathla and Gautam 2013.                                             clearly concentrated in a few states, but the
                                                                            trends in unorganized employment broadly
significantly higher shares of investment                                   track the size of the state. The exceptions
(about 31 percent), output (25 percent), and                                are Bihar, Madhya Pradesh, and Rajasthan,
employment (21 percent). As with the organized                              where both organized and unorganized food
segment, significant change has occurred across                             processing is underrepresented. On the other
states over time (Annex 13, Table A.13.1b).                                 hand, informality is the preferred mode of
                                                                            manufacturing in Karnataka, Odisha, and
The diversity of the organized and unorganized                              West Bengal. The broad correlation between
food-processing segments across states is                                   unorganized food processing and poverty likely

Figure 77: State poverty rates and shares in population, employment
    18                                                                                                                          70

    16
                                                                                                                                60
    14
                                                                                                                                50
    12

    10                                                                                                                          40

        8                                                                                                                       30
        6
                                                                                                                                20
        4
                                                                                                                                10
        2

        0                                                                                                                       0
              UP   MP   BH    WB       AP    MH     TN    RJ    KT     GJ     OR   KE   JH    AS    PB   CH      HY   UK   HP
                                Population        Unorganized        Organized     Rural Poverty HCR (2004/05)

Source: Bathla and Gautam 2013.




  146
                                                                                  Chapter 13: Agro-Industry: The Food-processing Sector




Figure 78: 
           Rural-urban ratios in the unorganized                                     Scale, factor intensity, and employment
           sector                                                                    creation in the food industry
  3.93
                 3.59                                                                Organized food processing has grown rapidly
                                                                                     since 1980 (Table 25), with capital deepening in
                                                                                     both the organized and unorganized segments.
                                                                                     Capital stock grew at 8 percent annually
                            0.51           0.46
                                                                                     between 1980 and 2008 in the organized
                                                       0.31        0.34
                                                                                     segment, primarily due to an increase in the
                                                                                     scale of operation (output per factory); the
   Enterprises




                 Workers




                            GVA/Worker




                                            GVA/Ent.




                                                       Inv./Ent.




                                                                   Capital Int.
                                                                                     number of factories grew at a much slower
                                                                                     pace. Employment per factory declined, but
                                                                                     total employment increased with the increase
Source: Bathla and Gautam 2013.                                                      in the number of factories. Between 1980
                                                                                     and 2008, the sector generated jobs at a rate
reflects a combination of “pull” and “push”                                          of 1 percent a year, with the fastest pace
factors, which call for deeper analysis.                                             between 2000 and 2008 at 2.3 percent per year.
                                                                                     In the unorganized segment, the number of
The final aspect of the sector’s structure is                                        enterprises has declined over time (Table 26),
the rural-urban dispersion of unorganized                                            but the remaining enterprises are operating at
enterprises (data to make this distinction                                           a larger scale, with rapidly growing output per
are not available for the organized segment).                                        unit and slower but still growing employment
Food-processing enterprises and employment                                           per enterprise. Overall employment in the
are almost four times more common in rural                                           unorganized segment fell due to the high exit
than in urban areas (Figure 78). The rural firms                                     rate of enterprises.
are much less capital intensive, and also less
productive both in output per enterprise and                                         The rapid increase in the average scale of
per labor. Apart from the greater dominance of                                       operation, visible in several states, was
OAMEs in rural areas, these ratios likely reflect                                    realized mostly in the post-reform period and
inadequate access to infrastructure, technology,                                     is consistent with the significant increase in
skills, and capital.                                                                 investment in the organized and unorganized

Table 25: Organized food-processing segment: Size and growth
                                                                                                      GVA              Capital stock
Size                       Factories (units)            Employment (millions)                (Rs billion, 2004/05) (Rs billion, 2004/05)
2008-09                        27,218                           1.56                                 352.5                 11,363
Growth (%/yr)                 Factories                Employment/factory                        GVA/factory       Capital stock/factory
1980-1989                         0.8%                         -3.5%                                  8.3%                  12.7%
1990-1999                                2.1%                             -0.1%                     5.1%                    8.6%
2000-2008                                1.8%                              0.5%                     6.4%                    5.4%
1980-2008                                1.6%                             -0.6%                     4.2%                    8.3%
Source: Bathla and Gautam 2013.




                                                                                                                                   147
Republic of India: Accelerating Agricultural Productivity Growth




Table 26:  Unorganized food-processing segment: Size and growth
                        Factories        Employment                              GVA                                                              Capital stock
Size                     (units)           (millions)                   (Rs billion, 2004/05)                                                 (Rs billion, 2004/05)
2005/06                    2.6                6.35                               142.5                                                                11,363
Growth (%/yr)          Enterprises       Employment/                      GVA/enterprise                                                         Capital stock/
                                          enterprise                                                                                               enterprise
2000/01–2005/06           -2.87%             1.38%                                               7.25%                                                3.85%
Source: Bathla and Gautam 2013.


segments. In the organized segment, for which      Figure 79: Trends in labor intensity
the temporal data are available, investments         2
increased substantially in the post-reform         1.8
period. Capital invested per factory grew          1.6
more than fourfold after 2000/02, when             1.4
investment grew by almost 20 percent per year.     1.2
Employment grew more slowly, although some           1
states saw a rapid expansion of employment         0.8
in the latest period (2000–08), including          0.6
Gujarat, Himachal Pradesh, Madhya Pradesh,         0.4

and Rajasthan. Employment per enterprise           0.2
                                                     0
increased slightly in the unorganized segment,
                                                         1980-81
                                                                   1982-83
                                                                             1984-85
                                                                                       1986-87
                                                                                                  1988-89
                                                                                                            1990-91
                                                                                                                      1992-93
                                                                                                                                1994-95
                                                                                                                                          1996-97
                                                                                                                                                    1998-99
                                                                                                                                                              2000-01
                                                                                                                                                                        2002-03
                                                                                                                                                                                  2004-05
                                                                                                                                                                                            2006-07
                                                                                                                                                                                                      2008-09
but the decline in the number of enterprises,
particularly in urban areas, caused employment
in the unorganized food-processing segment                                       Food Processing                                                    Non-Food Processing
overall to decline—a trend consistent with
                                                   Source: Bathla and Gautam 2013.
the rising capital intensity in the urban
unorganized food-processing segment.
Employment grew in only four states (Assam,        has almost converged, but significant gaps
Gujarat, Haryana, and Madhya Pradesh).             remain in wage rates and labor productivity in
                                                   favor of the non-food-processing industry.
Underlying these trends is a rapid decline
in labor intensity, with food processing
                                                   Factors determining employment (labor
experiencing a more rapid decline than non-
                                                   demand) in food manufacturing
food industries (Figure 79). Consistent with
this trend, labor productivity has risen, as       A statistical analysis of labor demand confirms
have wages, while capital productivity has         that increasing capital (and capital intensity)
declined. Labor productivity has risen faster      and scale of operation are associated with
in the non-food-processing sector, but wages       positive but low growth in employment
have increased more rapidly in the food            (Bathla and Gautam 2013). The elasticity of
sector. Figure 80 shows indicators for the food-   employment per factory with respect to scale
processing industry as a ratio of the indicators   (output per factory) is estimated to be 0.47,
for the non-food industry. Capital productivity    indicating that employment grows half as fast



  148
                                                               Chapter 13: Agro-Industry: The Food-processing Sector




           Ratios of indicators in food to non-
Figure 80:                                                       whereas most others showed varying degrees
           food-processing industry                               of increase. The overall trend in employment
                                                                  shows no consistent relationship, suggesting
                                                        0.95      that a mix of “pull” and “push” factors is at
                                                                  work across states. A deeper analysis of the
                        0.64
                                                 0.69             dynamics of the rural informal sector is
   0.57
          0.51                           0.53                     needed to disentangle the factors driving
                 0.47
                                                                  employment there.
                                  0.34

                                                                  In sum, although the Indian food-processing
                                                                  industry is more labor intensive than other
    Labor         Wages            Capital        Capital         industries, labor intensity has declined
 Productivity                     Intensity     Productivity      significantly as the industry has shifted toward
                        1980-85   2004-08                         more capital-intensive (labor-saving) technology.
Source: Bathla and Gautam 2013.                                   The good news is that labor productivity has
                                                                  risen fast, keeping pace with labor productivity
                                                                  in the non-food industry. The more rapid rise in
as output per factory in the organized segment.
                                                                  wages (relative to the non-food sector) has had
The elasticity for the unorganized segment is
                                                                  a dampening effect on employment, however,
also positive but lower at 0.17. As expected,
                                                                  encouraging further capital intensity and scale
wage rate has a strong negative impact on labor
                                                                  of operating units. Rising profits are inducing
demand, but that impact is much higher in the
                                                                  additional investment and the adoption of
organized segment than in the unorganized
                                                                  better technology, as expected, but enterprises
segment (–0.47 versus –0.07). Rising wages
                                                                  clearly prefer to move toward higher capital
induce substitution of capital for labor, as
                                                                  intensity and labor-saving technology. This
confirmed through an alternative specification
                                                                  preference reflects the perpetuation of
with capital intensity, which shows a negative
                                                                  informality in the manufacturing sector and a
and significant impact on employment in both
                                                                  reluctance to hire labor, given labor laws and
segments.
                                                                  other factors in the enabling environment.164
The unorganized food segment has lower
                                                                  The significance of these trends for employment
employment elasticity but is also becoming
                                                                  and transformation, even in the more rural food-
more capital intensive (albeit at a much
                                                                  based industry and in the more populous and
slower pace than organized processing). Labor
                                                                  poorer states, cannot be overstated. Encouraging
productivity has accordingly risen (with a
                                                                  new businesses to enter and existing businesses
high correlation between capital intensity
                                                                  to expand employment more rapidly than in
and labor productivity of 0.78 for 2005/06).
                                                                  the past will create employment, but only with
Wages have not risen as fast, probably because
                                                                  greater attention to the enabling environment
of the large labor pool in rural areas where
                                                                  and barriers to entry for smaller firms, especially
most unorganized food processing is located.
                                                                  in the lagging states.
Large variations across states prevail, however.
Capital intensity declined in some states, such                   164	 See the extensive discussion in the Economic Survey (GOI
as Gujarat, Himachal Pradesh, and Odisha,                              2013a) and World Bank (2012).




                                                                                                                            149
Republic of India: Accelerating Agricultural Productivity Growth




Patterns and Drivers of Private                                    regional distribution approaches that of the
Investment in Food Manufacturing                                   national economy. An RDI that is greater (less)
                                                                   than the national average indicates greater
Given the significant variation across states                      diversity (concentration).
in employment and private investment in the
food industry, it is important to understand                       The LQ and RDI results show significant
the patterns and drivers of investment. Based                      variation across states, summarized in
on standard metrics, this section assesses the                     Table 27 (for detailed state estimates, see
concentration and determinants of investment                       Annex 13, Table A.13.3a–b).167 Food processing
across states. Investment by a firm is based on                    has a greater presence (higher share) in the
many factors, such as profit (rate of return),                     organized segment of the industry in about
sales, liquidity, cost, and infrastructure.                        half (8 of 17) states in the 2000s, with some
Agglomeration economies, captured mainly                           changes over time. The most significant shifts
through location and urbanization effects, may                     have been a large decline in concentration in
loom large in firms’ investment decisions. At                      Assam and a large increase in Uttar Pradesh
the same time, higher productivity may also                        in the 2000s. Most states have remained
be achieved by greater diversity economies in                      either stable or increased, with consistent
locations having favorable policies for private                    and notable increases in Bihar, Kerala, and
investment (World Bank 2003).                                      Madhya Pradesh. In the unorganized segment,
                                                                   the share of food processing is high across a
Regional specialization and                                        much larger number of states (13). The level
concentration/diversification in private                           of representation is higher in urban areas,
investment                                                         both in terms of a higher number of states
                                                                   (15) and a higher LQ in most states relative
Two widely used two indices for specialization                     to rural enterprises. The states with higher
and concentration are the Location Quotient                        shares of investment (representation) in the
(LQ) and Relative Diversity Index (RDI). LQ                        food industry also have a relatively higher
shows the degree of representation of an                           share of output of the food industry in total
industry in a particular region/state relative to                  manufacturing.
other regions/states, with an LQ greater than 1
(less than 1) indicating higher representation                     The RDI estimates indicate that very few states
(dispersion) of an industry in the state relative                  have diversified their organized industrial
to the national average.165 RDI measures the                       base over time. Even in the 2000s, only four
degree of industrial concentration within                          states (Haryana, Karnataka, Maharashtra, and
individual states.166 As RDI increases, the                        West Bengal) have a high level of diversification
                                                                   (RDI greater than the national average at 0.95),
165	 LQ_ir is defined as LQ_ir = (Investment_is/Investment_s)/     with a very high degree of concentration in
     Investment_in/Investment_n), where Inv_is= Investment
     in industry i in state s; Inv_s= Investment in state s;
     Inv_in= Investment in industry i in country n; and Inv_n =    167	 LQ and RDI are calculated using average investment in the
     Investment in country n.                                           food industry drawn from ASI data at the two-digit level
166	 RDI is the inverse of the summed difference between the            classification for the organized segment in each decade from
     regional and national industrial shares, i.e.,                     1980–81 to 1989–90, 1990–91 to 1999–2000, and 2000–01 to
	    RDI_s =1/Σi|(Investment_is/Investment_s) – (Investment_in/)        2008–09. For the unorganized segment, the data are from the
     Investment_n|.                                                     two NSS rounds (2000–01 and 2005–06).




  150
                                                     Chapter 13: Agro-Industry: The Food-processing Sector




Table 27:  Classification of states as per LQ and RDI estimates based on investment
                                                                       States with RDI > national average
States with LQ > 1 (representation/concentration)
                                                                                   (diversity)
Organized segment, 1980–1989
Andhra Pradesh, Assam, Karnataka, Maharashtra, Punjab, Tamil         Andhra Pradesh, Gujarat,
Nadu, Uttar Pradesh,                                                 Maharashtra, Tamil Nadu
Organized segment, 1990–1999
Andhra Pradesh, Assam, Bihar, Karnataka, Kerala, Maharashtra,        Haryana, Madhya Pradesh,
Punjab, Tamil Nadu, Uttar Pradesh                                    Maharashtra, Tamil Nadu
Organized segment, 2000–2009
Andhra Pradesh, Assam, Bihar, Kerala, Madhya Pradesh, Punjab,        Haryana, Karnataka, Maharashtra,
Uttar Pradesh,                                                       West Bengal
Unorganized segment, 2000–2001
Andhra Pradesh, Assam, Bihar, Chhattisgarh, Haryana, Himachal
                                                                     Haryana, Karnataka, Kerala, Punjab,
Pradesh, Jharkhand, Kerala, Madhya Pradesh, Odisha, Rajasthan,
                                                                     Rajasthan
Uttaranchal, Uttar Pradesh, West Bengal
Unorganized segment, 2005–2006
Andhra Pradesh, Assam, Bihar, Himachal Pradesh, Jharkhand,           Haryana, Karnataka, Madhya Pradesh,
Karnataka, Kerala, Madhya Pradesh, Odisha, Uttar Pradesh,            Punjab, Rajasthan, Tamil Nadu,
Uttaranchal, West Bengal                                             West Bengal
Source:  Bathla and Gautam 2013.


most states. In the unorganized segment,                is dominant in every state, whether industrially
overall diversification is low (with a national         advanced or less developed.
average of only 0.22). Urban enterprises are
significantly more diversified, but with the            Second, food manufacturing is more dominant in
majority of food enterprises located in rural           states with a higher percentage of poor people.
areas, the total RDI reflects the concentration         Individuals in the agricultural sector apparently
of rural enterprises. Relative to the national          diversify out of agriculture by starting small
average, about half of the states are more              nonagricultural enterprises or operating such
diversified (8) in the unorganized segment.             enterprises alongside their agricultural ventures
                                                        to supplement their incomes.
Three main findings emerge from the analysis.
First, in the organized segment, the food               Third, industrial investments are highly
and beverage industry has relatively higher             concentrated in specific sectors, with the
representation (based on LQ) in states that             majority of states having an RDI less than
are less industrially developed and have                the national average. The weak link between
a relatively higher share of income from                diversity and performance (growth in output
agriculture. In contrast, the spatial distribution      per enterprise) suggests that beyond the
and locational representation of the food and           benefits of competition, other factors, including
beverage industry in the unorganized segment            agglomeration economies (specialization in



                                                                                                      151
Republic of India: Accelerating Agricultural Productivity Growth




certain subsectors or industries) and especially                     urban demand, and others), the LQ for factories
linkages back to agriculture (for access to raw                      is used as a proxy.
materials and cost considerations), may be
important reasons for the private sector to                          The descriptive statistics for the data used in
invest in food processing in certain states.                         the analysis are given in Annex 13, Table A.13.4.
                                                                     Investment rate (investment per capital stock)
Patterns of private investment                                       varies between 10 and 17 percent across states,
                                                                     but investment rate as a ratio of value added
The role of different factors in explaining inter-                   is significantly higher and also considerably
state variations in the pattern of investment in                     more variable. Most states also show a high rate
food manufacturing is assessed using a flexible                      of return on capital. As noted, capital output
accelerator model (see Bathla and Gautam                             ratios are low across the states, as is capacity
2013). In addition to the economic variables                         utilization. It is also clear that the food industry
and industry characteristics common to such                          depends heavily on borrowed capital, with a
investment functions, key variables of interest                      high ratio of indebtedness (outstanding loan to
are the policy variables, namely infrastructure,                     capital) in most states. Across states, the rate
agricultural linkage, and investor friendliness.                     of investment in food processing is relatively
Given India’s diverse agro-climatic conditions,                      higher in agriculturally dominant states and
which favor the cultivation of many different                        states with relatively lower per capita incomes.
crops, agricultural linkages are likely to be                        The exceptions are Punjab and Haryana, which
important for locating investments. Two
                                                                     have a higher share of agricultural income and
alternative variables, size of the agricultural
                                                                     higher annual per capita income.
economy (share of agricultural state GDP in
total state GDP) and land productivity (defined
                                                                     Results of the econometric analysis of the
as agricultural state domestic product per unit
                                                                     determinants of investment show significant
of net sown area) are used to test this linkage.
                                                                     impacts of internal industry factors as well
Capital expenditures by state governments
                                                                     as external (that is, state-specific) factors
to build roads and bridges are used to test
                                                                     (see Annex 13, Table A.13.5). As expected,
the “crowding in” effect of infrastructure
                                                                     investment responds to profits and sales
investments on private investment in food
                                                                     (accelerator). Borrowing as a ratio of capital is
processing.168 The proxy for an investor-friendly
                                                                     negative, suggesting possible credit constraints
environment is the number of industrial
                                                                     to expansion, but it is only weakly significant
strikes and lockouts in a state.169 To capture
                                                                     and is sensitive to specification.
any other location effects (broad agglomeration
economies such as skilled labor force, size of
                                                                     Private investment is also influenced by other
                                                                     factors. Infrastructure development is an
168	 A weighted average of public capital expenditures on roads
     and bridges, on a per capita basis, for the previous four       important determinant of investment. Clearly,
     years is used to assess the crowding-in effect. This approach   higher public investment in roads and bridges
     makes it possible to capture the lags in the likely impact of
     infrastructure development on investors’ expectations and       has a strong “crowding-in” effect on private
     response.                                                       investment. Backward linkage of the food
169	 The impact of market reforms as initiated under the
     APMC Model Act could not be gauged because most states
                                                                     industry to agriculture, captured through the
     implemented the reforms only recently (Chapter 11).             share of agricultural income in total state



  152
                                                                   Chapter 13: Agro-Industry: The Food-processing Sector




income, is positive and significant, suggesting                       including subsidies, single-window clearance
potential economies of scale and accessibility to                     mechanisms, tax holidays, and the mobilization
raw material as strong determinants. Further,                         of required resources. To complement these
an alternative specification using agricultural                       initiatives, creating the enabling environment
productivity also shows a strong positive                             for private investment by removing the
impact, indicating that not only size but also                        regulatory and institutional constraints (such as
higher agricultural productivity can stimulate                        the marketing issues discussed in the previous
investment in food processing.170 The estimated                       chapter) needs to be a high priority. The recent
elasticity for public infrastructure is substantial                   opening up of FDI in retail is likely to yield
at 0.24, but the elasticity for linkage to                            significant benefits for those states that move
agriculture is even higher at 0.99. Agricultural                      most rapidly to initiate and implement investor-
productivity shows an elasticity of 0.74. Finally,                    friendly reforms in the agricultural sector.
investor friendliness has an expected negative
sign but is not statistically significant, although
this result is likely clouded by the state fixed                      Productivity Growth, Technical
and lagged investment effects.                                        Change, and Technical Efficiency
                                                                      in the Food-processing Industry
The key results from a policy perspective are
                                                                      Trends in partial factor productivities show that
the backward linkages to agriculture, the
                                                                      labor productivity has grown consistently in
importance of infrastructure, and the role
                                                                      the organized segment of the food-processing
of credit. Backward linkages to agriculture
                                                                      industry since the 1980s, while capital
are clearly important for increasing private
                                                                      intensity increased rapidly (though both have
investment (and thus employment) through
                                                                      slowed from their very high rates of growth
food processing. The food industry’s
                                                                      in the 1980s) (Annex 13, Table A.13.6). Capital
concentration in agricultural states and
                                                                      productivity has declined for the most part,
relatively lesser presence in the more industrial
                                                                      although modest growth began in the 2000s.
states such as Gujarat and Maharashtra is
                                                                      The unorganized segment shows similar trends,
consistent with this result and clearly suggests
                                                                      despite the relatively short time between the
that “location” matters.
                                                                      two surveys in the 2000s, with a slightly higher
                                                                      annual growth rate in capital productivity than
Encouraging private investment in food
                                                                      in the organized segment. Interestingly, the
processing in the relatively more agricultural
                                                                      fairly high correlation between capital intensity
states such as Bihar, Himachal Pradesh,
                                                                      and labor productivity across states in the 1980s
Madhya Pradesh, and Uttarakhand could
                                                                      (0.78) declined in subsequent decades (to 0.22 in
contribute significantly to poverty reduction.
                                                                      the 2000s), showing no clear pattern in any state
Recently several states have sought to attract
                                                                      except Maharashtra (especially in the 2000s).
private investment through incentives

                                                                      Given these divergent trends, assessing the
170	 The specification uses agricultural productivity from the
     previous year (in other words, a one-year lag) to avoid the      performance of the food-processing industry
     problem of potential endogeneity with current investment.        using total factor productivity (TFP) gives a
     The impact of infrastructure becomes insignificant in the
     specification with agricultural productivity as the two
                                                                      better picture of how well the industry is
     variables are correlated.                                        performing. Estimates of the annual TFP growth



                                                                                                                  153
 Republic of India: Accelerating Agricultural Productivity Growth




             FPG in organized and unorganized food
 Figure 81: T                                                                              variability, but on average TFPG is estimated
            manufacturing, 2000–09 and 2000–05                                             to be higher for rural enterprises. The
                           4                                        Odisha                 correlation between TFPG for the organized and
                                                                         RAJ
                                                                                           unorganized segment across states is strong and
TFPG Unorganised 2000-05




                                                          Gujarat KAR MP
                           2
                                                          UP
                                                              Haryana Punjab               positive (Figure 81), indicating that states doing
                                      Assam
                                                                MH                         better in one segment also do better in the
                           0                                         India                 other, though at very different rates of growth.
                                                                       TN
                                                     HP     Bihar                          The implication is that some states may have
                           -2                                 AP
                                                                                           a better overall investment climate for food
                                                   Kerala
                                                                                           processing.
                           -4
                                                                                 WB
                                -10           -5            0                5        10   The better growth in TFP for the organized food
                                              TFPG Organized 2000-08                       industry may be attributed to the acquisition
 Source: Bathla and Gautam 2013.                                                           of better technology, driven by the growing
                                                                                           demand for processed food and a policy
 (TFPG) are given in Annex 13, Table A.13.7.                                               change in favor of this segment. The low
 Figure 81 gives the scatter plot of TFPG in both                                          growth in TFP in unorganized food processing
 the segments during 2000–08.                                                              is consistent with the weak position of many
                                                                                           micro-enterprises, reflected in their high exit
 For all of India, TFPG was positive in the pre-                                           rates, low capacity utilization, and inadequate
 reform period (1.28 percent), declined in the                                             resources and/or skills. These problems are
 immediate post-reform period of the 1990s                                                 particularly acute among the OAMEs that
 (–0.45 percent), and picked up in the 2000s                                               represent a large share of the enterprises in the
 to attain its highest level (1.76 percent).                                               unorganized segment.
 The decline in the 1990s possibly reflects
 the increased investments in capital, while                                               The unorganized segment might grow through
 labor growth continued at a steady but lower                                              a complementary relationship with organized
 rate. This pattern is consistent with lags in                                             food manufacturing, yet while TPFG is positively
 the benefits from capital investment, which                                               correlated between the two segments, in the
 perhaps started to be realized in the 2000s. Like                                         2000s the correlation was low and statistically
 the other indicators, TFPG varies across states. A                                        insignificant (0.28). The correlation between
 majority of states saw a rise in TFPG after 2000,                                         TFPG in the organized food segment and growth
 except for Assam, Bihar, Himachal Pradesh,                                                in GVA in the unorganized segment is also low,
 Kerala, and Uttar Pradesh.                                                                indicating weak complementarity (for example,
                                                                                           through outsourcing or subcontracting)
 The data make it possible to estimate TFP for                                             between the two segments.
 the unorganized segment only for the post-
 reform period. TFPG is estimated separately                                               Another important finding is the growing
 for rural, urban, and all areas. At the aggregate                                         relationship between TFPG and growth in
 level, TFPG for the unorganized segment was                                               output (GVA) in both the organized (correlation
 positive but much lower than in the organized                                             of 0.63) and unorganized (correlation of 0.77)
 segment. Across states there is considerable                                              segments, indicating growing efficiency in the



                           154
                                                                      Chapter 13: Agro-Industry: The Food-processing Sector




Figure 82: Relationship between growth in GVA and TFP
                                               MP                        30
15                                                            Haryana                                                                 HP

                                     Punjab                   MH         20
                                                       KAR
10            HP                                  India UP
                                                           AP                                            RAJ
                                                   TN Kerala             10              Orissa      HR     KAR MH
                                                    RAJ                                         Ker     PNIndia
                                                    Gujarat WB                                  TN  Guj          Bihar
 5                                                                                                      AP UP
                                                               Bihar
                                                                          0                  MP WB
                                    Odisha
 0                                                                      -10
     -10                  -5                  0                   5           -10            -5                    0             5                 10

                          TFP Growth 1980-89                                                             TFP Growth 1990-99
                   95% CI                  Fitted values                                          95% CI                  Fitted values
                   GVA Growth 1980-89                                                             GVA Growth 1990-99


                                               MP   RAJ                  20
15                                                                                                                               KAR
                                          KAR          WB                                                                        Haryana
                                    Haryana      Punjab
10                                           Odisha                                                                        Assam Gujarat
                               HP       Guj                              10
                                            India                                        Kerala                             UP     RAJ
                                  UP AP TN                                                               TN                       MP
 5                                                                                                                 India                       Odisha
           Assam                                                                                                                 Punjab
                                      MH                                                          AP                  MH
                                                                          0                            Bihar
                          Kerala     Bihar
 0
                                                                                WB
                                                                                                        HP
-5                                                                      -10
     -10             -5             0               5            10                 -4       -2                0             2             4
                           TFP Growth 2000-08                                              TFP Growth Unorganised 2000-05
                   95% CI                     Fitted values                                95% CI                 Fitted values
                   GVA Growth 2000-08                                                      GVA Growth Unorganised 2000-05

Source: Bathla and Gautam 2013.


use of inputs. This relationship has evolved                             estimates the relative technical efficiency in
over time. Figure 82 plots GVA and TFP growth                            food manufacturing to assess the extent of
in food processing in major states for each                              inefficiency in the industry—either because
of the three decades from 1980–81 onward.                                subsidies and other incentives drive inefficient
No relationship existed in the 1980s, but a                              input use or because of other factors.
positive relationship developed in the 1990s                             Technical efficiency is a key component of
and remained positive in the 2000s. The same                             TFP and provides an assessment of the relative
relationship holds for the unorganized segment                           performance of individual states.171
in the 2000s.

                                                                         171	 The analysis uses stochastic frontier analysis based on time
Explaining TFP and technical efficiency                                       invariant and time varying decay inefficiency models. The
in food processing                                                            estimates are based on an estimated standard Cobb-Douglas
                                                                              production function with labor and capital (fixed capital
                                                                              stock from ASI data and value of fixed assets from the NSS
What explains changes in productivity                                         data for the unorganized segment). See Bathla and Gautam
growth in food manufacturing? This analysis                                   (2013) for details.




                                                                                                                                               155
Republic of India: Accelerating Agricultural Productivity Growth




The estimated factor elasticities indicate                             Figure 83: Technical efficiency across states
scope for additional labor absorption                                  100%
through expansion, with significantly higher                           90%
                                                                                All India Average
potential in the unorganized segment.172 In                            80%
the unorganized segment, rural enterprises                             70%
                                                                       60%
tend to be more productive than others.
                                                                       50%
Finally, both the organized and unorganized
                                                                       40%
segments seem to be operating below optimal                            30%
scales—both exhibit increasing returns to scale.                       20%
The fact that enterprises in the unorganized                           10%
segment operate well below optimal levels                               0%
highlighted throughout India, as labor laws                                   OR WB AP UP KE AS GJ TN BH KT MH HP MP RJ HY Pb
and other regulatory constraints create                                Source: Bathla and Gautam 2013.
strong disincentives for enterprises to expand
(GOI 2013a).
                                                                       efficiency in the 1980s, which they maintained
                                                                       in subsequent decades (Annex 13, Table A.13.8),
In terms of technical efficiency, the average
                                                                       whereas other states started at relatively lower
firm appears to operate at about 77 percent
                                                                       levels and have fallen further behind, with
efficiency, although levels vary widely across
                                                                       consistent declines over the decades.
states, from 56 percent in Odisha to 95 percent
in Punjab. Several states are at the high end of
                                                                       These results demonstrate the considerable
the efficiency scale, including Haryana, Madhya
                                                                       scope to improve productivity in the organized
Pradesh, and Rajasthan in addition to Punjab,
                                                                       food industry in most states. This goal could
implying that the potential for easier gains
                                                                       be achieved in agriculturally dominant states
through pure technical efficiency appears to be
                                                                       through appropriate measures to exploit
lower than in other states. At the other end of
                                                                       their comparative advantage in offering
the scale, Andhra Pradesh, Kerala, Odisha, Uttar
                                                                       lower transaction and transport costs (for
Pradesh, and West Bengal are operating at low
                                                                       raw materials), better economies of scale, and
levels of efficiency, suggesting significant scope
                                                                       cheap labor. A favorable policy of appropriate
for improvement.
                                                                       incentives and enabling environment offers
                                                                       promise, as shown by the examples of Himachal
Figure 83 shows average efficiency levels for the
                                                                       Pradesh and Madhya Pradesh (and lately
entire period (1980–2009), but efficiency levels
                                                                       Uttarakhand), which offered incentives for
have varied in most states over time.173 Haryana,
                                                                       private investment. Himachal Pradesh has a
Himachal Pradesh, Karnataka, Madhya Pradesh,
                                                                       small share in the food industry in terms of
Punjab, and Rajasthan had high levels of
                                                                       factories, workers, investment, and output, but
                                                                       efficiency and productivity are much higher
172	 The coefficient of labor, estimated at 0.61, is slightly higher   there than in other states.
     than the coefficient for capital at 0.48 in the organized food
     industry. In contrast, the value of coefficients of labor in
     NDME-DME is much higher at 1.56 and lower for capital             What explains the variation in the food
     at 0.37.
173	 Decadal estimates are derived from a time varying decay
                                                                       industry’s productivity from one state to
     inefficiency model.                                               the next? This question has received less



  156
                                                      Chapter 13: Agro-Industry: The Food-processing Sector




attention than warranted by India’s spatial              processing. A 1 percent increase in share of
heterogeneity. It can be examined by                     agriculture in the state economy is associated
estimating the effects of the following factors          with an increase in the productivity of the
on TFP (subject to the availability of data):            food processing sector by about 0.3 percent. As
infrastructural development, of which two                expected, investor friendliness is important,
particularly important public investments                with strikes and lockouts having a significant
are roads (represented by road density) and              negative impact on productivity. Urbanization
power (measured as electricity consumption               is another significant determinant of
per capita). With respect to power, an                   productivity in food processing, reinforcing the
equally important dimension for investing in             importance of access to markets and consumer
processing plants (as in manufacturing more              demand. The LQ for the industry is positive
broadly) is access to power, which may be                but not statistically significant, suggesting
limited given the share of electricity used by           that agglomeration economies appear to
agriculture (often heavily subsidized). Other            be adequately captured through backward
factors include investor friendliness (numbers           linkages, urbanization, and institutional
of strikes and lockouts as a proxy for the               environment.
investment environment); localization index
(measured as the LQ for factories to test for
agglomeration economies); agricultural linkages
                                                         Summary and Implications
(size of agriculture in the state economy to             Geographical and agricultural conditions
capture the importance of backward linkages              heavily influence an environment’s
in determining the productivity of the food              favorability for improving productivity in
industry); and urbanization (to proxy market or          food manufacturing, encouraging additional
demand side factors).                                    investment in the industry, and creating much-
                                                         needed jobs of good quality off of the farm.
The results ( Annex 13, Table A.13.9) confirm            This result is understandable, given the benefits
the earlier finding of strong cross-state                to be had from good access to raw materials
differences as captured by the state fixed               with fewer transaction and transport costs. It is
effects, clearly showing that the broader                consistent with and helps provide a rationale
policy and institutional environment matters             for the earlier findings that private investment
significantly for productivity. These strong             in food processing tends to be concentrated in
state-specific effects, which reflect public             more agricultural states.
investment decisions, appear to cloud the
effect of roads. Controlling for state fixed             In addition to the agro-ecological endowments
effects, the results suggest that improvements           and structural factors, the results show that
in road density over time within a state (the            state-specific factors such as the policy and
“within” state effect) have a positive impact,           institutional environment play a very large
although it is not as strong statistically as other      role in explaining TFP and productivity growth.
results. The agricultural linkages again have            These factors are not fully understood, and they
a high and significant impact, showing that              could not be captured in the analysis above
the relative size of the agricultural economy            using the data in hand. Further research will
helps improve the productivity of food                   be critical to expand the understanding of the



                                                                                                     157
Republic of India: Accelerating Agricultural Productivity Growth




role of policies that can support growth in        unorganized segment of the industry, which
productivity in food processing and agriculture    is more widespread, has higher potential to
more generally.                                    absorb labor, but has low productivity. Public
                                                   policy thus needs to focus on promoting growth
The findings highlighted here point to             in agricultural productivity and increasing
the significant potential for agriculturally       investments in supportive infrastructure such
dominant states to attract private investment      as irrigation, roads, and marketing facilities,
in the food industry. Rapid expansion of this      within a climate that will attract private
relatively more labor-intensive industry in        investment. The high levels of (technical)
laggard states with low per capita income,         efficiency within the food-processing sector in
high dependence on agriculture, and high           many states suggests that the food industry is
incidence of poverty will help to absorb more      well placed to compete in a more liberalized
people from agriculture. These arguments           marketplace, suggesting potential scope for FDI
are even more relevant with respect to the         as well as exports.




 158
   Transforming Agriculture in East India
E. 




G
        eographical diversification of the                             agricultural productivity. At the same
        production base from the traditional                           time, they house the largest number of
        grain basket (the irrigated upper                              undernourished and poor people in India
Gangetic plains) is a high priority for the                            (more than 40 percent). Multiple reasons have
government. An important element of the                                been forwarded for the poor performance and
broader strategy is to promote growth in                               historically poor agricultural growth of these
agricultural productivity in the eastern                               regions.176 They have a high risk of concurrent
states, given the sustainability concerns and                          floods and droughts, small and fragmented
diminishing returns in cereals in northwestern                         landholdings, little infrastructure to facilitate
India. With substantial water, a suitable climate                      agriculture, weak institutions, poor governance,
for cereals (especially rice), and significant                         and a poor policy response to the changing
room to improve yields, the underdeveloped                             needs of agriculture. The general perception
eastern states are an attractive alternative for                       is that eastern India presents enormous
sustaining national food security. The focus on                        opportunities if its natural resources can be
catalyzing the transformation process in the LIS                       judiciously managed, appropriate policies
and rainfed areas—which occupy 60 percent                              and institutions established, and supportive
of the country’s area—is also driven by their                          agricultural infrastructure put in place. With
potential to produce high-value commodities                            the limited resources available, it is important
such as oilseeds, pulses, horticultural crops, and                     to focus and sequence any reforms and
livestock.174                                                          interventions in these areas to ensure that they
                                                                       realize their potential for improved agricultural
Eastern and Central India (comprising Bihar,                           growth. This chapter conducts an in-depth
Uttar Pradesh, Jharkhand, Madhya Pradesh,                              assessment of two states—Bihar and Odisha—to
Odisha, and Chhattisgarh) are well endowed                             gain a better understanding of the constraints
with natural resources for agriculture,175 yet                         and opportunities involved in promoting more
they are food-deficit areas with very low                              rapid agricultural growth in eastern India.




174	 LIS include Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh,
     Odisha, Rajasthan, and Uttar Pradesh.
175	 Northeastern states, including Assam, are classified as Special
     Category States, and are not the focus of the analysis here.      176	 NAAS (2010, 2011).




                                                                                                                   159
14                     Priorities for Agricultural
                       Growth in Bihar and Odisha




T
       he Low-income States (LIS)177 have long               both groups experienced a similar reduction in
       lagged behind the other states in overall             poverty in recent years.179 Within the LIS and
       and agricultural growth. Their GDP                    non-LIS, states with a relatively larger share of
growth improved significantly between 2005                   the economy in agriculture also had a lower
and 2012, narrowing the differential with the                head-count ratio of poverty in 2011–12 (Annex
non-LIS, but still remains lower on average at               14, Figure A.14.3). Drawing causal inferences
7.8 percent per year compared to 8.5 percent                 from these trends may be misleading (a deeper
all other states (Annex 14, Figure A.14.1). The              analysis of the drivers of poverty is required),
average for the four eastern states (Bihar,                  but they do not prima facie suggest inherent
Chhattisgarh, Jharkhand, and Odisha) is                      links between poverty and the share of
about the same as the LIS average. In contrast,              agriculture in the economy.
agricultural growth in the LIS (4.9 percent) has
been higher than the non-LIS (about 3 percent),              India’s broader development strategies focus
suggesting convergence, although the LIS                     on agriculture as an important component
started from a very low base. The eastern states’            of making growth more pro-poor. The RKVY
average agricultural growth is close to the                  “Bringing Green Revolution to Eastern India”
overall LIS rate.                                            program, for example, focuses explicitly on
                                                             promoting agricultural production in eastern
The level of poverty reflects the level of                   India, to promote inclusive growth as well as
development in each state (Annex 14,                         maintain food security in light of the concerns
Figure A.14.2). The seven LIS states collectively            about the sustainability of production in the
house 62 percent of India’s poor and 64 percent              traditional northwestern breadbasket states.180
of the rural poor.178 Poverty is overwhelmingly              Recent experience from Gujarat shows that
rural, averaging 84 percent across the LIS                   appropriate policies and interventions can
(the national average for rural areas is about               stimulate agricultural productivity and growth
80 percent), with the highest incidence in                   (Shah et al. 2009).181
Odisha (91 percent) and Bihar (90 percent). With
the exception of Rajasthan, all rural poverty                179	 Poverty estimates on national poverty line using uniform
rates in LIS are higher than in non-LIS, although                 reference period, for 2004–05 (Planning Commission
                                                                  2013 data).
                                                             180	 For details, see http://bgrei-rkvy.nic.in/
177	 Lagging states include Assam, Chhattisgarh, Jharkand,   181	 Gujarat averaged annual agricultural growth of 9.6 percent
     Madhya Pradesh, Odisha, Rajasthan, and Uttar Pradesh.        between 2000 and 2007. Gujarat’s growth has slowed since
178	 Estimated from Planning Commission (2013) data.              2007 but remains on par with that of other states.




                                                                                                                       161
Republic of India: Accelerating Agricultural Productivity Growth




Figure 84: Change in labor and land productivity for LIS and non-LIS, 1990–2008
.00006
                                                                     .00004

.00004                                                               .00003

                                                                     .00002
.00002
                                                                     .00001


     0                                                                     0
         0       10000      20000       30000       40000    50000           0      20000     40000     60000     80000 100000
   Labor Productivity (Real Agricultural Value of Output/Labor)       Land Productivity (Real Agricultural Value of Output per Hectare)

                                                                                             1990 Non-LIS       1990 LIS
                         1990 Non-LIS           1990 LIS                                     2008 Non-LIS       2008 LIS
                         2008 Non-LIS           2008 LIS

Source: Authors’ estimates, using district agricultural productivity data from Kumar and Jain 2012.


This chapter examines the barriers to                                Agriculture’s share in economic output is
agricultural growth in Bihar and Odisha, two of                      falling much faster than the share of people
the poorest and least developed eastern states,                      employed in agriculture, which remains
and identifies opportunities for faster and more                     disproportionately large at 63 percent of Bihar’s
inclusive growth. Given the heterogeneity and                        10.9 million workers and 62 percent of Odisha’s
potential governance issues (since agriculture is                    16.2 million workers in 2009.183 Continued
a state responsibility), a case study approach is                    rapid population growth and insufficient exit of
adopted for this analysis.182 After characterizing                   labor out of agriculture has resulted in falling
the current status of agriculture in these two                       labor productivity, even as land productivity
states, the following sections discuss the specific                  improved between 1990 and 2008. This
challenges to the sector in each state and                           situation contrasts sharply with that of non-LIS
potential interventions to overcome them.                            (Figure 84).

                                                                     Recent years show some improvement,
Agricultural Performance                                             with Bihar and Odisha experiencing higher
Agriculture’s share in the economies of both                         agricultural growth rates than the non-LIS after
Bihar and Odisha has declined quite rapidly;                         performing poorly in the 1990s (Bihar grew
in 2011–12, agriculture and allied sectors                           slower than the national average; Odisha had
accounted for 22 percent and 16 percent                              negative growth). Despite the good performance
of their respective state economies (GSDP),                          in the last decade, significant scope for
respectively (Table A.14.1). Both states face a                      improvement remains in both states, as the
significant structural transformation problem:                       following sections indicate.

                                                                     183	 These shares are estimated from Planning Commission
182	 The Bihar- and Odisha-specific details in subsequent sections        (2013) data and are lower than the shares in 2004. Since
     draw heavily from preliminary drafts of background papers            then, employment growth has been faster in nonagricultural
     prepared by IFPRI (2012a, 2012b).                                    sectors, such as industry in Bihar.




  162
                                                Chapter 14: Priorities for Agricultural Growth in Bihar and Odisha




Low yields and wide yield gaps                                                  Average yields of major crops in
                                                                     Figure 85: 
                                                                                selected states (2009/10, MT/ha)
Very low yields are typical of agriculture in both
                                                                                     5
states (Figure 85), but large yield differentials
for crops and livestock are found across                                             4




                                                                    Yield (ton/ha)
districts within each state (IFPRI 2012a, 2012b).
                                                                                     3
Despite generally favorable growing conditions
(plenty of water, sunlight, and good soils), the                                     2
substantial variation in local agro-ecologies calls
for a localized rather than generic approach.                                        1

For example, the most popular crop in both                                           0
districts is rice, but yields on nearly half                                                    Paddy                 Wheat                   Maize                    Pulses
(47 percent) of Bihar’s rice area are less than                                           Odisha         Bihar       Punjab        Haryana              West Bengal              India
1.5 tons per hectare. Similarly, yields in one-                      Source: 
                                                                             MoA (various), Directorate of Statistics and
third of the wheat area are less than 2 tons per                             Evaluation, Department of Agriculture,
hectare. Maize has performed better in Bihar,                                Government of Bihar.
with relatively fast growth in output, half of it
due to improved yields.                                                         Rice yield gaps in Bihar and Odisha,
                                                                     Figure 86: 
                                                                                2009 (difference between district
                                                                                average yields and frontline
Rice is even more important in Odisha,                                          demonstration yields, by variety)
occupying about half of the gross cropped
                                                                                     6
area. While the area under rice has been
steady, its share in gross cropped area has                                          5
slowly declined.184 Rice yields grew reasonably
                                                                    Yield (t/ha)




                                                                                     4
                                                                                                                                                         1.9                     2.1
well after 2001 (from 1.1 tons per hectare in                                        3                                            3.5
                                                                                                                                                                     0.8
                                                                                                                                              4.2
2000/01 to 1.6 tons per hectare in 2008/09),                                                           1.6
                                                                                     2     1.3
for an annual increase in rice production                                                                                                                              3         2.7
                                                                                     1                                                                   2.6
of 6.47 percent. Almost all of this increase                                               1.4         1.5    1.2
                                                                                                                       0.9        1.2
                                                                                                                                              0.7
(97 percent) was from yield improvements.                                            0
                                                                                                              -0.1
                                                                                          Rajendra
                                                                                           Kasturi
                                                                                                       Rajendra
                                                                                                       Suwasini
                                                                                                      Improved
                                                                                                         Samba
                                                                                                        Mahsuri
                                                                                                                       Sampada

                                                                                                                                  Rajendra
                                                                                                                                 Mahsuri-1
                                                                                                                                             Rajendra
                                                                                                                                             Bhagwati

                                                                                                                                                         Dhanrasi

                                                                                                                                                                        Satya
                                                                                                                                                                    Krishana

                                                                                                                                                                                 Chandan
                                                                                     -1
Yield gaps are large. In rice, yield gaps for
popular varieties are as great as 300 percent
in Bihar and 80 percent in Odisha (Figure 86).                                                                  Bihar                                Odisha
                                                                                                     Average District Yield          Difference from FLD Yield
Substantial yield gaps remain in maize in Bihar,
despite rapid recent growth, calling for better                      Source: DRR and IFPRI 2012a, 2012b.
management practices.                                                Note: Bihar districts are Bhagalpur, Munger, Banka,
                                                                     Nawada, Jahanabad, Patna, Nalanda, Samastipur
                                                                     and Muzaffarpur; Odisha districts are Dhenkanal,
The LIS generally have significantly lower milk                      Jagatsinghpur, Jajpur and Puri.
and livestock productivity (Leitch, Ahuja, and
Jabbar 2013). For example, milk and overall                          Haryana are at least double the levels in
livestock productivity in Punjab, Kerala, and                        Bihar and Odisha. IFPRI (2012a, 2012b) finds
                                                                     substantial gaps in milk yields in Bihar and
184	 Share of area under rice declined from 56 percent in 2000–01
     to 48 percent in 2009–10.                                       Odisha (Table 28).



                                                                                                                                                                                163
Republic of India: Accelerating Agricultural Productivity Growth




           Milk yields in Bihar and Odisha (kg/day/
Table 28:                                                                                                                                                             percent and livestock for 30 percent of the
           per animal) in 2009–10                                                                                                                                      value of agricultural output in 1990–91. By
                                                                       Average                             Yield range under                                           2008–09, their shares had risen to 23 percent
                                                                        yield                                 best practice                                            and 40 percent, respectively. In contrast, the
   Bihar                                                                                                                                                               share of rice was only 11 percent and wheat was
   Crossbred cattle                                                            6.2                                    9.5–11.1                                         8 percent in 2008–09.
   Indigenous cattle                                                           2.9                                     4.2–4.9
   Buffalo                                                                     3.9                                     5.5–5.9                                         Similar patterns are observed in Odisha.
   Odisha                                                                                                                                                              The share of high-value commodities (fruits,
   Crossbred cattle                                                            5.9                                         10–15                                       vegetables, livestock, and fisheries) in output
                                                                                                                                                                       has risen rapidly. Fruits and vegetables are the
   Indigenous cattle                                                           1.2                                        3.7–5.7
                                                                                                                                                                       largest share of output, with eggplant, tomato,
   Buffalo                                                                     2.9                                           6–8
Source:  Basic Animal Husbandry Statistics 2010, GOI.
                                                                                                                                                                       cabbage, cauliflower, and okra being the most
                                                                                                                                                                       important. They accounted for 65 percent of
                                                                                                                                                                       vegetable production and 56 percent of area
A rapidly diversifying production base
                                                                                                                                                                       under cultivation during 2009–10.
As shown in Chapter 6, diversification drove
nearly one-third of the growth in the crop                                                                                                                             Despite their seeming agro-ecological
sector in East India between 2004 and 2009.                                                                                                                            similarities and increasing agricultural
Both Bihar and Odisha are good examples                                                                                                                                diversification, the two states show distinct
of a rapidly diversifying production base in                                                                                                                           patterns and sources of growth (Figure 87),
which the share of food grains has fallen. In                                                                                                                          which belie broad conclusions about growth
Bihar, fruits and vegetables accounted for 14                                                                                                                          processes across the subregion. For example,

            ources of growth in Bihar and Odisha, 1981–2010
Figure 87: S
       10                                                                                                                                                                         6
                                                                                       Bihar                                                                                                                                                                 Odisha
           8
                                                                                                                                                                                  4
           6

           4                                                                                                                                                                      2
Millions




                                                                                                                                                                       Millions




           2
                                                                                                                                                                                  0
           0

           -2                                                                                                                                                                     -2
                1981-82
                          1983-84
                                     1985-86
                                               1987-88
                                                         1989-90
                                                                   1991-92
                                                                             1993-94
                                                                                       1995-96
                                                                                                 1997-98
                                                                                                           1999-00
                                                                                                                     2001-02
                                                                                                                               2003-04
                                                                                                                                         2005-06
                                                                                                                                                   2007-08
                                                                                                                                                             2009-10




           -4
                                                                                                                                                                                  -4
                                                                                                                                                                                       1981-82
                                                                                                                                                                                                 1983-84
                                                                                                                                                                                                           1985-86
                                                                                                                                                                                                                     1987-88
                                                                                                                                                                                                                               1989-90
                                                                                                                                                                                                                                         1991-92
                                                                                                                                                                                                                                                   1993-94
                                                                                                                                                                                                                                                             1995-96
                                                                                                                                                                                                                                                                       1997-98
                                                                                                                                                                                                                                                                                 1999-00
                                                                                                                                                                                                                                                                                           2001-02
                                                                                                                                                                                                                                                                                                     2003-04
                                                                                                                                                                                                                                                                                                               2005-06
                                                                                                                                                                                                                                                                                                                         2007-08
                                                                                                                                                                                                                                                                                                                                   2009-10




      -6
                                    Area                 Yield                 Price                   Diver                   Interaction
                                                                                                                                                                                  -6
                                                                                                                                                                                                            Area                Yield                 Price                Diver                Interaction
Source: Authors, using data from Birthal et al. 2014.




           164
                                    Chapter 14: Priorities for Agricultural Growth in Bihar and Odisha




Bihar has seen strong and consistent               marketable surplus is small, transaction costs
diversification, in which prices have played a     and market risks tend to be very high.
lesser role. An important finding is that land
is increasingly being taken out of production,     About 90 percent of the holdings in Bihar
which is undoubtedly a drag on growth.             are under 1 hectare (72 percent are under 0.5
Improved yield spurred growth in Odisha in the     hectare). The average land holding is extremely
early 2000s, but now growth is led primarily       small in Bihar (0.43 hectare in 2005–06)
by prices. The rising contribution of prices       (Annex 14, Table A.14.2). It is higher in Odisha
and declining contribution of yields raises        (1.15 hectare in 2006–07), but farm sizes are
concerns about the sustainability of growth.       declining there as well. Already 60 percent of
In comparison, the sources of growth in Bihar      holdings in Odisha are less than 1 hectare. The
appear to be more robust.                          small size of holdings, absentee landlordism,
                                                   and tenancy and sharecropping arrangements
Significant diversification has occurred beyond    discourage investments in agriculture,
the crop subsector with a rising contribution      especially for land development and irrigation.
from the livestock sector over the past two        Tenancy is pervasive in Odisha, mostly on a
decades, (Annex 14, Table A.14.3) (IFPRI           sharecropping (cash or kind) basis (IFPRI 2012a).
2012a, 2012b). Leitch, Ahuja, and Jabbar
(2013) estimate that in 2008/09, the livestock     Growth in productivity is sluggish
subsector was responsible for 69 percent of
agricultural growth in Bihar and 42 percent in     As at the national level, over the long run
Odisha. These contributions are substantially      (1980–2008) crop production (excluding
higher than the 36 percent share of livestock in   horticultural crops) has been heavily input-
agricultural growth for all of India, indicating   driven (Figure 88). Bihar and Odisha were
the importance of livestock as part of the
growth strategy, given rapidly rising demand                  Sources of output growth in selected
                                                   Figure 88: 
for livestock products and the limited land for               states, 1981–2008 (%)
extensive cultivation of crops.                     4.00

                                                                              2.86
                                                    3.00
Challenges                                                                                           2.25   2.35

                                                    2.00
Holdings are small and fragmented                           1.44
                                                                     0.99
Smallholders predominate in Bihar and               1.00                                   -0.78
Odisha. The fragmentation of land holdings
is an additional challenge to efficiency and        0.00
productivity in Bihar. The “nano” scale of                 Assam     Bihar &   Madhya Odisha        West    India
                                                                   Jharkhand Pradesh &             Bengal
operations is a constraint in marketing output     -1.00                     Chhatisgarh
(especially bulky, low-value cereals) and in
accessing information on grades and sanitary       -2.00              Input          TFP           Output

and phytosanitary standards necessary to move      Source: 
                                                           Calculated using estimates of Kumar, Gautam, and
up the value chain (IFPRI 2012b). When the                 Joshi 2013.




                                                                                                            165
Republic of India: Accelerating Agricultural Productivity Growth




among the states with the slowest growth in                                                              Table 29: Fertilizer use in Bihar and Odisha
production; production actually contracted
                                                                                                                                                   Growth in
in Odisha in the 1990s. While Bihar’s crop                                                                                                       NPK intensity
sector experienced some TFP growth, Odisha’s                                                                         NPK use intensity (kg/ha)    (% per year)
hemorrhaged as productivity fell throughout                                                                            2000          2008           2001–08
the 1980s and 1990s (Figure 89). Odisha’s TFP                                                            Bihar         95.76        174.04            10
has also been highly erratic over the past two                                                           Odisha        40.49          58.96             6
decades. Between 1981 and 1995, TFP declined
                                                                                                         India         89.73        127.67              5
at an average annual rate of 2.5 percent, but it
                                                                                                         Source: Estimated from CoC data.
has since risen faster than in any other state
(1.86 percent).
                                                                                                         been half of the all-India average (Table 29),
Low input use efficiency is a major contributor                                                          though it has been growing in recent years. In
to sluggish and negative productivity growth                                                             contrast, Bihar’s fertilizer use is well above the
in Bihar and Odisha, partly because seed                                                                 Indian average and grew at an average annual
replacement rates are low and fertilizer use is                                                          rate of 10 percent in 2001–08, faster than the
imbalanced. In Bihar, the seed replacement rate                                                          national average. In 2009, Bihar’s fertilizer
has hardly ever exceeded 15 percent, except for                                                          use intensity was 231 kilograms per hectare,
maize, rapeseed-mustard, rice, and wheat; the                                                            with fertilizer use in some districts rivaling
highest replacement rate is about 30 percent                                                             that in Punjab (400 kilograms per hectare)
for wheat. In Odisha, the seed replacement rate                                                          or Haryana (320 kilograms per hectare). As
is less than 14 percent. Seed replacement rates                                                          elsewhere in India, in Bihar fertilizer use is
are higher where hybrids are available, such as                                                          imbalanced. More urea (nitrogen) is applied
hybrid sunflower (about 83 percent).                                                                     relative to other nutrients (phosphorus and
                                                                                                         potassium). Fertilizer use in Odisha is within
Fertilizer use is more differentiated. Odisha’s                                                          the recommended range.
fertilizer intensity (NPK kg/ha) has consistently
                                                                                                         High risk of biotic and abiotic stress
           TFP in traditional crops, Bihar and
Figure 89: 
                                                                                                         Multiple biotic and abiotic stresses, often
           Odisha, 1981–2008
                                                                                                         at high levels, add significantly to the risks
160                                                                                                      inherent in agricultural production in both
140                                                                                                      states. They are a critical constraint on the
120                                                                                                      adoption of new technology, the efficiency of
100                                                                                                      input use, and productive investments, limiting
                                                                                                         potential growth. Quantitative data are not
80
                                                                                                         available on the nature and extent of losses
60
                                                                                                         from biotic and abiotic stresses, but focus group
40                                                                                                       discussions identified a number of stress factors
      1981

             1983

                    1985

                           1987

                                   1989

                                          1991

                                                 1993

                                                        1995

                                                               1997

                                                                      1999

                                                                             2001

                                                                                    2003

                                                                                           2005

                                                                                                  2007




                                                                                                         affecting crops.
                                  Bihar & Jharkhand                          Odisha

Source: 
        Estimated from CoC data (Kumar, Gautam, and                                                      In Bihar for example, rice is affected by sheath
        Joshi 2013).                                                                                     blight, bacterial leaf blight, and blast. Stem



  166
                                       Chapter 14: Priorities for Agricultural Growth in Bihar and Odisha




borers can lead to losses ranging from 25–40           300,000 hectares, soil acidity affects 236,000
percent of the final output of rice, maize and         hectares (25.1 percent of cropped area), and
vegetables (IFPRI 2012b). Similar risks were           waterlogging affects 628,000 hectares. Odisha’s
identified for wheat (rust) and pulses (pod            abiotic constraints include floods, cyclones,
borers).                                               droughts, acidity and water logging (Annex 14,
                                                       Table A.14.5). Deficiencies of zinc, boron, and
Odisha also faces a number of biotic constraints       sulfur reduce yields, especially in upland areas.
exacerbated by high rainfall and humidity,
which are conducive to a range of crop diseases        Poorly functioning markets
and pests (see Annex 14, Table A.14.4 for some
of the pests and diseases and their associated         Marketing remains a challenge in both states,
risks) (IFPRI 2012a). Crop damage from pests           where commodity prices often fall below prices
and diseases ranges from 20 percent for the            farmers obtain elsewhere in India. A useful
most widely grown rice variety to 70 percent in        benchmark is the government-established MSP,
maize. The susceptibility to multiple diseases         which is often the floor price for areas that are
and pests makes production very risky.                 major producers of commodities supported
                                                       through public procurement. A comparison
Abiotic stresses are common in both states,            for Bihar is given in Table 30. Farmers in
including frequent and multiple threats of             Odisha similarly receive rice prices that are
floods, drought, soil problems (acidity and            10–25 percent lower than the MSP, even though
sodicity),185 and waterlogging (IFPRI 2012a,           Odisha’s rice markets are well integrated (IFPRI
2012b). Floods are a major challenge for Bihar,        2012a). Odisha’s vegetable markets, on the
where nearly 41 percent of cropped area                other hand, are not so well integrated, and
(2.2 million hectares) is prone to flooding. Bihar     prices vary significantly across markets.
experienced severe floods in 2004/05, 2007/08,
2008/09, and 2010/11. In the past five years,          Markets in Bihar and Odisha tend to be thinly
droughts have also occurred almost every year.         spread and underdeveloped. The low density
Sodicity adversely affects crop yields on about        of markets raises transportation costs from the

           Farm harvest prices (FHP) of paddy, wheat, and maize in Bihar and their Minimum Support
Table 30:  
           Price (MSP) (Rs/quintal)
                               Paddy                                  Wheat                 Maize
                       MSP                    FHP
                Grade A  Common      Winter         Rainy      MSP            FHP     MSP           FHP
2001–02           560      530        418            363        620           517     485           366
2002–03           580      550        411            355        630           539     490           397
2003–04           580      550        410            361        630           566     505           403
2006–07           650      620        497            455        850           817     540           514
2007–08           775      745        633            549      1,000           938     620           575
Source: IFPRI 2012b.

185	 Sodium content.




                                                                                                     167
Republic of India: Accelerating Agricultural Productivity Growth




farm to primary or secondary markets. Markets                        facilities are virtually absent. As markets play
often have insufficient basic infrastructure                         no active role in price realization, nonmarket
and facilities. Only a very limited number have                      forces determine the eventual prices.
invested in more sophisticated infrastructure
such as cold storage.                                                Infrastructure gaps
Stakeholders highlighted several constraints                         Infrastructure shortfalls in Bihar and Odisha are
in marketing agricultural commodities (IFPRI                         clear barriers to growth. Poor infrastructure has
2012a, 2012b). With the repeal of the APM(D&R)                       far-reaching effects on agricultural productivity
Act, Bihar has no regulatory framework for                           by raising input prices, reducing output prices,
markets. Any private player can set up a                             increasing volatility through limiting market
market without government clearance, but                             integration, and constraining the development
private investments have not taken place due                         of value chains and value addition. Bihar and
to other constraints, such as the lack of access                     Odisha, like other eastern states, face a number
to land—including the now-defunct APM(D&R)                           problems related to infrastructure, partly
markets. Private traders have taken over the                         because past public investments have been low.
marketplaces amid allegations of collusion                           As noted in Chapter 11 and Annex 11 (Figure
and monopolistic/monopsonistic behavior                              A.11.1), eastern India as a region continued to
by traders/agents, who exact fees as before                          receive far less capital investment per capita
but do not invest in market infrastructure or                        for irrigation and infrastructure (roads and
provide any services.186 The situation is viewed                     electricity) than other regions (Singh 2012). The
as untenable and unconducive to orderly                              various Five Year Plans also show that Bihar and
and transparent marketing or proper price                            Odisha’s allocation of plan resources (in rupees
discovery.                                                           per person) has been consistently lower than
                                                                     the all-India average since the very first plan.
In Odisha, the Act is intact, but most markets                       In the Eleventh Five Year Plan, Bihar received
have no mechanism for price discovery                                Rs. 6,576 per capita and Odisha received
and price determination. Prices are mostly                           Rs. 8,205 per capita, versus an all-India average
“negotiated” rather than determined by an                            of Rs. 13,187. This under-allocation is at odds
orderly process. The lack of transparency                            with the plan’s stated objectives of pro-poor and
prompts frequent complaints of farmers being                         inclusive growth.
“cheated” by traders.
                                                                     Despite this resource constraint, Odisha and
Despite the range of vegetables grown in                             Bihar have narrowed the gap in road density
Odisha, the market infrastructure is generally                       with the rest of India (on average). Odisha’s
negligible. A few markets provide basic                              road density is 138 kilometers per thousand
infrastructure such as a market yard and                             square kilometers (km/thousand km2); it also
cleaning and sanitation facilities. Storage                          has a long seacoast (with three ports) and a
186	 These allegations, voiced through stakeholder focus group
                                                                     network of railways. Bihar has more than
     discussions, need to be investigated formally through a         1,276 km/thousand km2—higher than the all-
     detailed market study on the current (post-repeal) structure,
     conduct, and performance of agricultural markets in Bihar,
                                                                     India average (812 km/thousand km2). Current
     including the development of farm and wholesale prices.         road density is a significant improvement over



  168
                                      Chapter 14: Priorities for Agricultural Growth in Bihar and Odisha




2003, when the state government started to                       Extent of irrigation and sources of
                                                      Figure 90: 
invest heavily in roads. Road density increased                  irrigation water in Bihar and Odisha,
                                                                 2008 (%)
at about 9.8 percent per year between 2003
and 2008. On a per capita basis, however,             70.0

Bihar lags behind the national average. Road          60.0
density in Bihar is 1.3 km per thousand persons,      50.0
compared to the all-India average of 3.7 km per       40.0
thousand persons.
                                                      30.0

                                                      20.0
Other infrastructure bottlenecks remain more
                                                      10.0
binding, particularly rural electrification
(IFPRI 2012a, 2012b). By 2008, only                    0.0




                                                                                                                      share of irrigated
                                                                                                                          area supplied
                                                                                                                           by tubewells
                                                                                              share of irrigated
                                                                                                  area supplied
                                                                                                      by canals
52 percent of inhabited villages in Bihar were




                                                                   share of net
                                                                 sown area that
                                                                    is irrigated
electrified—60 percent in Odisha—in contrast
to 100 percent in Andhra Pradesh, Haryana, and
Punjab; 98 percent in Karnataka; and the all-
India average of 81 percent. In Odisha, electricity          Bihar                 India- irrigation share         India-canal share
consumption in agriculture is only 1.5 GWh per               Odisha                India-tubewell share
100,000 persons, among the lowest in India,           Source: Authors, using DES, MOA data.
where the average is about 45 GWh per 100,000
persons. While electricity consumption in
agriculture has increased over the years in other     for groundwater are diesel-operated and thus
states, in Odisha it declined from 1.7 GWh per        more expensive to operate than electric-
100,000 persons in 2007–08 to 1.5 in 2008–09.         powered pumps, but that consideration has
                                                      not prevented irrigation from spreading (IFPRI
The lack of electricity prevents the LIS in           2012a). Bihar could improve its irrigation
general from increasing irrigation intensity          system by better exploiting its surface water
(Annex 14, Figure A.14.4), but Bihar and Odisha       potential.
appear to be outliers in terms of irrigation
development. Bihar, with 62 percent of its            Odisha’s irrigation profile is the opposite of
net sown area irrigated, surpasses the all-India      Bihar’s, with greater reliance on surface water
average of 45 percent, while Odisha has 38            irrigation: 65 percent of irrigation water is from
percent of its area equipped for irrigation. The      canals and only 15 percent is sourced from
nature of irrigation in these two states explains     groundwater. This difference is partly a result
how this development occurred despite severely        of Odisha’s geological profile, which in many
limited access to electricity.                        places restricts access to groundwater.

Groundwater provides 64 percent of irrigation
                                                      Priority Areas for Growth
water in Bihar (Figure 90), while 28 percent
is sourced from canals, despite access to             Overcoming the challenges to agricultural
three major rivers—the Ganges, Kosi, and              growth in Bihar and Odisha will require
Gandak. Moreover, 92 percent of the pumps             resources and investments—making it



                                                                                                                                      169
Republic of India: Accelerating Agricultural Productivity Growth




necessary to prioritize agricultural investments   likely soon face the same problem. In Bihar, the
in overall public expenditures in addition to      predominance small farms mean that prospects
reorienting current public expenditures to         for improving livelihoods and prosperity
achieve the maximum impact. Historically           among those remaining in farming are likely
agricultural investments have been low in          to be enhanced by diversifying into higher-
both states. Within the existing resource          value crops in Bihar. Odisha still has significant
envelope, Bihar’s agriculture received about       potential in traditional crop production, but
4 percent of the Tenth Five Year Plan outlay,      diversification will remain an integral part of
while irrigation and flood control received        sustainable productivity growth and poverty
about 11 percent. The allocation of these          reduction.
expenditures prioritized infrastructure for
food grain storage and crop husbandry, which       To raise productivity, the high risks from biotic
together accounted for 54 percent of the total     and abiotic stresses need to be mitigated. For
expenditure. The livestock sector contributes      both states, better land use through sustainable
about 40 percent of agricultural value added but   multiple cropping holds significant promise in
received an allocation of only about 10 percent.   this regard. Both have witnessed considerable
An even smaller proportion was spent on dairy      diversification already, and significant scope
development. For Odisha, the share of public       remains for further gains. Technology and
expenditure in Agriculture and Allied Activities   innovation can enable the current rice-
in total public expenditure was 4 percent in       fallow systems to make better use of land
2009–10, a marked improvement from the             and irrigation. Exploiting options other than
0.8 percent of 2006–07.                            rice (such as legumes, as in the Barind Tract
                                                   of Bangladesh), in combination with short-
The three main areas for intervention to           duration rice varieties, will improve the use of
enhance agricultural growth in Bihar and           soil moisture as well soil fertility and farmers’
Odisha are improving productivity, marketing,      incomes.187
and livestock sector development. These are
discussed next.                                    Technological solutions include relatively
                                                   simple strategies, such as ensuring good quality
                                                   seed and increasing seed replacement rates.
Strategies and technologies to                     Research and extension can make it easier to
mitigate risks and increase                        cope with abiotic stresses, as in the case of the
productivity                                       flood-tolerant rice varieties (like the Swarna-
Bihar and Odisha have significant opportunities    sub1) that are now becoming popular in
to improve agriculture’s performance in ways       Odisha and the hybrids being grown in Bihar.
that maximize its capacity to reduce poverty.      Promoting and partnering with private seed
The prospects for each state lie along similar
but not identical paths. The small size of         187	 Nearly 2 million hectares in Bihar and 1.5 million hectares
                                                        in Odisha are left fallow during the rainy season to preserve
holdings in Bihar make it more imperative to            soil moisture for the subsequent rabi (spring) season. New
focus on enabling labor to exit agriculture by          technologies are available to improve the use of land and
                                                        water resources without compromising soil moisture for rabi
developing skills and creating nonfarm jobs.            crops, and they will significantly improve land productivity
Odisha is not in the same situation but will            and farmers’ incomes.




 170
                                     Chapter 14: Priorities for Agricultural Growth in Bihar and Odisha




companies to propagate hybrid seed, as for          levels in the 1990s (Annex 14, Table A.14.6).188
maize in Bihar, is a high priority. Biotechnology   Since 2000, research spending has increased at
is another important and unexploited tool           11 percent and 15 percent per year in Bihar and
to improve stress resistance and tolerance in       Odisha, respectively, but spending on extension
new varieties and hybrids, enabling farmers to      has stagnated. ATMA and KVKs are the two
increase productivity. Improved management          important channels connecting innovation
practices to mitigate crop risks are equally        and farmers, and coordination between these
important.                                          components of the system is necessary to reap
                                                    the benefits of improved technologies.
Input use and efficiency can increase
significantly, especially for fertilizer and        A focus on linking smallholders to
irrigation. In Odisha the major constraints         remunerative markets
to using more fertilizer are soil acidity and
sodicity. Both have relatively simple technical     Marketing interventions need to focus on
solutions that can significantly improve crop       smallholders to link them with input and
response to fertilizer and irrigation. Through      output markets. In this respect, collective
new programs (or perhaps through reorienting        action—for example, through producer
current public programs such as RKVY),              organizations, cooperatives, farmer associations,
acidic and sodic soils could be reclaimed and       or self-help groups, as well as contract farming,
improved.                                           farmer companies, and farmer clubs—is
                                                    important to achieve scale economies. Bihar
Improved management practices have                  already has several successful approaches that
enormous potential, given that the region’s         could be scaled up, such as the Sudha program
abundant water could be better managed              for dairy production and marketing, and the
to reduces losses and increase production.          Samridhi program through the Kaushalya
Surface water management and drainage must          Foundation to integrate vegetable producers
receive high priority to direct excess water        and shorten the supply chain in Nalanda
                                                    District. These innovations offer important
from flood-prone areas to drought-prone areas.
                                                    lessons on the importance of using sustainable
A comprehensive irrigation and drainage
                                                    farmer institutions to realize the benefits of
program, including hydrological and technical
                                                    new opportunities.
feasibility assessments, needs to evaluate the
options for better water management (such
                                                    Besides mobilization at the farm level, measures
as linking rivers with adequate provision of
                                                    are needed to link farmers with remunerative
drainage) (IFPRI 2012a).
                                                    markets and value chains. Such measures could
                                                    include evolving special agriculture zones in
Finally, agricultural research and development
                                                    niche areas such as cereal, fruit, vegetable,
need to be reoriented, in addition to receiving
                                                    poultry, fish, or milk production and processing;
better support from a robust extension
program. Spending on agricultural research          188	 In Bihar, 2.7 percent of public expenditure was allocated to
and extension was low in Bihar and Odisha;               agricultural research, compared to 0.1 percent in Odisha
                                                         (IFPRI 2012a, 2012b). Agricultural research spending was
extension has been generally low for decades,            Rs. 6.2 billion in 2007 for India as whole, Rs. 289 million in
but spending on research declined to negligible          Bihar, and Rs. 169 million in Odisha.




                                                                                                                  171
Republic of India: Accelerating Agricultural Productivity Growth




initiating contract farming; creating an             Growth in the livestock subsector—
enabling environment and incentives to attract       a “quick win”
private investment in markets, food processing,
and land development; and reforming the              For both Bihar and Odisha, growth in the
policy and regulatory framework to reduce            livestock subsector represents a relatively
market-related risks, promote transparency, and      “quick win” in terms of poverty impact,
allow the free movement of goods and services.       empowering women, improving nutritional
                                                     outcomes, and generating employment for
An example of a program to link producers            the landless or for marginal farmers. The
with remunerative markets is development of          subsector’s strong prospects for growth are seen
Krushak Bazaars (farmers’ markets) in Odisha.        in its rapid transformation throughout India,
To promote direct links between farmers and          with rising demand for higher-value animal
consumers, the Government of Odisha has              products and significant potential for exports
established 43 Krushak Bazaars since 2000,           (Chapter 10). For example, rapidly expanding
through incentives including the provision           exports of buffalo meat constitute an important
of land and infrastructure. Wholesale prices         opportunity for Bihar. Livestock accounts for
in Krushak Bazaars are higher than in the            an increasing share of agricultural output in
wholesale markets but lower than retail prices       each state.
(based on primary data collected by IFPRI 2012b),
ensuring that farmers and consumers gain. There      Even so, the subsector remains vastly
is scope for improvement: pricing rarely involves    underdeveloped for milk, meat, and other
farmers, and participating farmers find that the     livestock products. Yield levels remain very
prices do not account for quality differences.       low, indicating the substantial scope for
There is also evidence that these markets are        improvement through better service provision.
dominated by nonfarmers. Involving farmers           Veterinary services, breeding facilities, access
in determining prices, and creating adequate         to good quality feed/fodder, and processing
infrastructure, are two areas for improvement.       facilities are all in very short supply. Yields of
                                                     crossbred cows remain very low for several
Another option is to develop horticulture            reasons, including indiscriminate breeding,
clusters that focus aggressively on horticultural    low artificial insemination (AI) rates (Leitch,
commodities from production to marketing.            Ahuja, and Jabbar 2013), limited facilities
Initiatives in horticulture have been taken          for AI (IFPRI 2012a), feed constraints, and
up in an integrated manner by promoting              high morbidity because of the high disease
the cultivation of mangoes, bananas, spices,         burden. Priority actions to remedy these
and flowers. With a significant rise in the          problems include a reorientation of the
production of horticultural crops during the last    agricultural research agenda to focus on major
10 years, the Odisha government has decided          livestock diseases and production constraints,
to identify clusters by mapping areas producing      increased policy and public support for
specific varieties of fruits and vegetables. Along   small ruminant, pig, and poultry production
with mapping, the government is creating a           (with their significant potential for growth),
policy framework for the private sector to invest    and incentives for private investment in
in modern market infrastructure, particularly        processing, value chains, feed production, and
for horticultural crops.                             veterinary services.


  172
                  References




Background papers for this report                         2012b. “Transforming Bihar’s
                                                   __________.

                                                   Agriculture: Challenges and Opportunities.”
Ahuja, R. 2014. “Agriculture and Private Sector
Linkages.”                                         Kumar, A., and R. Jain. 2012. “District Level
                                                   Agricultural Productivity Temporal and Spatial
Babu, S., P. Joshi, C. Glendenning, K. Aseno-      Trends.”
Okyere, and R. Sulaiman. 2012. “The State of
Agricultural Extension and Research Reforms in     Kumar, P., M. Gautam, and P.K. Joshi. 2013.
India: Strategic Priorities and Policy Options.”   “Total Factor Productivity Growth: A State Level
                                                   Analysis.”
Bathla, S., and M. Gautam. 2013. “Accelerating
Private Investment and Competitiveness in          Kshirsagar, V., and M. Gautam. 2013.
Food Processing Industry in India: A State Level   “Agriculture Productivity Trends in India: A
Analysis.”                                         District Level Analysis.”

Bisaliah, S., and Dev. S.M. 2012. “Private         Lele, U., M. Agarwal, P. Timmer, and
Investments in Indian Agriculture: Farm Level      S. Goswami. 2013 “Patterns of Structural
Evidences and Policy Directions.”                  Transformation and Agricultural Productivity
                                                   Growth: With Special Focus on Brazil, China,
Chand, R., and S. Parappurathu. 2012.              Indonesia, and India.”
“Historical and Spatial Trends in Indian
                                                   Leitch, H., V. Ahuja, and M. Jabbar. 2013.
Agriculture: Growth Analysis at National and
                                                   “India’s Livestock Sector: Demand Growth, Food
State Level.”
                                                   Security and Public Investment – Issues and
Gautam, M., K. Pradhan, and H. Nagarajan. 2013.    Options.”
“Technical, Economic, and Allocative Efficiency
                                                   Nin-Pratt, A., and M. Gautam. 2013.
and Its Determinants of Indian Agricultural
                                                   “Productivity, Technical Change, and
Farmers Using ARIS/REDS Panel Data.”
                                                   Efficiency: A DEA Application to Indian Farm
Ganesh-Kumar, A. and M. Panda. 2013. “An           Households.”
Analysis of Input-Output Multipliers for India.”
                                                   Patnaik, G., and S.K. Sharma. 2013. “Agricultural
Ganguly. R. 2012. “Pro-Poor Value Chains.”         Marketing—Policy Framework.”
IFPRI (International Food Policy Research          Rawal, V., and M. Swaminathan. 2012.
Institute). 2012a. “Energizing Odisha’s            “Returns from Crop Cultivation and Scale of
Agriculture: Challenges and Opportunities.”        Production.”


                                                                                                  173
Republic of India: Accelerating Agricultural Productivity Growth




Reardon, T., B. Minten, R. Hernandez, C. Hu,           edited by P.K. Aggarwal. New Delhi: Indian
and S. DasGupta. 2012. “Analysis of Alternative        Council of Agricultural Research (ICAR).
Output and Input Marketing Arrangements and
                                                       Ahluwalia, M.S. 2011. “Prospects and Policy
Their Impacts on Farm-Level Technology Use
                                                       Challenges in the 12th Plan.” Economic and
and Productivity in India.”
                                                       Political Weekly, May 21.
Rodgers, G., and A.N. Sharma. 2013. “Structural
                                                       Ahuja, V. 2004. “Economic Rationale for Public-
Change in Bihar’s Agriculture: Findings from a
Longitudinal Study.”                                   Private Delivery of Livestock Services.” OIE
                                                       Scientific and Technical Review 23 (1): 33–45.
Singh, S. 2012. “Contract Farming for
Sustainable Agricultural Development in India:         Ahuja, V., M. Dhawan, M. Punjabi, and
A Smallholder Perspective.”                            L. Maarse. 2008. “Poultry-based Livelihoods
                                                       of the Rural Poor: Case of Kuroiler in West
                                                       Bengal.” Research Report. South Asia Pro-Poor
Other References                                       Livestock Policy Program, Food and Agriculture
                                                       Organization (FAO), New Delhi.
Acharya, S.S. 2004. Agricultural Marketing in India.
Vol. 17 of Millennium Study of Indian farmers.         Ahuja, V., K.E. McConnell, D. Umali Deininger,
New Delhi: Government of India, Academic               and C. de Haan. 2003. “Are the Poor Willing
Foundation.                                            to Pay for Livestock Services? Evidence from
                                                       Rural India.” Indian Journal of Agricultural
        2006. “Agricultural Marketing and Rural
__________.

Credit for Strengthening Indian Agriculture.”          Economics, 58 (1): 84–100.
INRM Policy Brief, Asian Development Bank,             Ahuja, V., J. Morrenhof, and A. Sen. 2003.
New Delhi.                                             “Veterinary Services Delivery and the Poor: Case
Acharya, S.S., R. Chand, P.S. Birthal, S. Kumar,       of Rural Odisha, India.” OIE Scientific and Technical
and D.S. Negi. 2012. “Market Integration               Review 22(3): 931–48.
and Price Transmission in India.” Food and             Ahuja, V., M. Rajasekhar, and R. Raju. 2008.
Agriculture Organization (FAO), New Delhi.             Animal health for poverty alleviation: A review
Ackerman, R. 2011. New directions for water            of key issues for India. Background paper
management in Indian agriculture. Processed.           prepared for the World Bank. Processed.

Aggarwal, P.K., N. Kalra, S. Chander, and H.           Ahuja, V., D. Umali-Deininger, and C. de Haan.
Pathak. 2006. “Info-Crop: A Dynamic Simulation         2003. “Market Structure and the Demand
Model for the Assessment of Crop Yields, Losses        for Veterinary Services in India.” Agricultural
Due to Pests, and Environmental Impact of              Economics 29: 27–42.
Agro-ecosystems in Tropical Environments. I.           Alston, J.M., C. Chan-Kang, M.C. Marra,
Model Description.” Agricultural Systems 89: 1–25.     P.G. Pardey, and T. J. Wyatt. 2000. “A Meta-
Aggarwal, P.K., and D.N.S. Rani. 2009.                 Analysis of the Rates of Return to Agricultural
“Assessment of Climate Change Impacts on               R&D: Ex Pede Herculem.” IFPRI Research Report
Wheat Production in India.” Chapter 2 in Global        No. 113. Washington, DC: International Food
Climate Change Impact and Indian Agriculture,          Policy Research Institute (IFPRI).




   174
                                                                                                   References




Anderson, J.R. 2007. Agricultural advisory             Intelligence” being carried out at Indian
services. Background paper for the World               Agricultural Statistics Research Institute (IASRI),
Development Report 2008, World Bank,                   New Delhi.
Washington, DC.
                                                       Bathla, S. 2011. Trade Liberalization and Indian
Anderson, K., M. Ivanic, and W. Martin. 2014.          Agriculture. Rawat Publications, Jaipur.
“Food Price Spikes, Price Insulation, and
                                                       Bathla, S., and R. Srinivasulu. 2011. “Price
Poverty.” In The Economics of Food Price Volatility,
                                                       Transmission and Asymmetry: An Empirical
edited by J.-P. Chavas, D. Hummels, and B.D.
                                                       Analysis of Indian Groundnut Seed and Oil
Wright. Chicago: University of Chicago Press.
                                                       Markets.” Indian Journal of Agricultural Economics,
Anderson, K., and S. Nelgen. 2012. “Updated            Oct-Dec.
National and Global Estimates of Distortions to
                                                       Beintema, N., G-J Stads, K. Fuglie, and P. Heisey.
Agricultural Incentives, 1955 to 2011.” Database.
                                                       2012. “ASTI Global Assessment of Agricultural
World Bank Agricultural Distortions Website.
                                                       R&D Spending: Developing Countries Accelerate
Badiani, R., and K.K. Jessoe. 2011. Electricity        Investment.” International Food Policy Research
subsidies for agriculture: Evaluating the impact       Institute (IFPRI), Washington, DC.
and persistence of these subsidies in India.
                                                       Bhagwati, J.N. 2002. “Democracy and
University of California, Berkeley. Processed.
                                                       Development: Cruel Dilemma or Symbiotic
Banerji, A., and J.V. Meenakshi. 2004. “Buyer          Relationship?” Review of Development Economics
Collusion and Efficiency of Government                 6(2): 151–62.
Intervention in Wheat Markets in Northern
                                                       Bhalla, G.S., and G. Singh. 2001. Indian
India: An Asymmetrical Structural Auction
                                                       Agriculture: Four Decades of Development.
Analysis.” American Journal of Agricultural
                                                       New Delhi: Sage Publications.
Economics 86(1): 236–53.
                                                               2010. “Growth of Indian Agriculture: A
                                                       __________.
Basu, K. 2011. “India’s Foodgrain Policy: An
                                                       District Level Study.” Final Report on Planning
Economic Theory Perspective.” Economic and
                                                       Commission Project, Planning Commission,
Political Weekly, January 29.
                                                       Government of India, New Delhi.
Basu, K., and A. Maetrens. 2007. “Pattern and
                                                       Bhavani, T.A., A. Gulati, and D. Roy. 2006.
Causes of Economic Growth in India.” Oxford
                                                       “Structure of the Indian Food Processing
Review of Economic Policy 23(2): 143–67.
                                                       Industry: Have Reforms Made a Difference?” In
Bathla, S. 2008. “Agriculture Reforms and              Plate to Plough: Agricultural Diversification and its
Market Integration: A Spatial Analysis of Food         Implications for the Small holders in India. Report
and Non-food Commodities.” Journal of Social and       submitted to the Ford Foundation by the
Economic Development 10(2): 196–221.                   International Food Policy Research Institute
                                                       (IFPRI), Washington, DC.
Bathla, S. 2009. “Market integration in
agriculture markets: an analysis of selected           Binswanger-Mkhize, H.P., and A. D’Souza.
cereals, oilseeds and horticultural crops.” Report     2011. “Structural Transformation of the Indian
submitted to National Agricultural Innovation          Economy and of Its Agriculture.” In Productivity
(NAIP) Project on “Policy Analysis and Market          Growth in Agriculture: An International Perspective,



                                                                                                        175
Republic of India: Accelerating Agricultural Productivity Growth




edited by K.O. Fuglie, S. Ling Wang, and V. Eldon              2012. “Diversification in Indian
                                                        __________.

Ball. Wallingford, UK: CAB International.               Agriculture Towards High Value Crops: The Role
PP. 181–98.                                             of Small Farmers.” Canadian Journal of Agricultural
                                                        Economics 61(1): 61–91.
Birner, R., S. Gupta, and N. Sharma. 2011.
“The Political Economy of Agricultural Policy           Birthal, P.S., and S. Kumar. 2009. “Conditions for
Reform in India: Fertilizers and Electricity for        the Success of Contract Farming in Smallholder
Irrigation.” International Food Policy Research         Agriculture?” Policy Brief No. 30. National
Institute (IFPRI), Washington, DC.                      Center for Economics and Policy Research
                                                        (NCAP), Indian Council of Agricultural Research
Birthal, P.S. 2008. “Linking Smallholder
                                                        (ICAR), New Delhi.
Livestock Producers to Markets: Issues and
Approaches.” Indian Journal of Agricultural             Birthal, P.S., and D.S. Negi. 2012. “Livestock for
Economics 63(1): 19–37.                                 Higher, Sustainable, and Inclusive Agricultural
                                                        Growth.” Economic and Political Weekly
Birthal P.S., A.K. Jha, and A.K. Joseph. 2006.
                                                        (Supplement) 47(26 & 27): 89–99.
Livestock production and the poor in India.
Capitalization of Livestock Program Experiences         Blümmel, M., and P.P. Rao. 2008. Fodder
in India (CLAPI), New Delhi. Processed.                 market in Hyderabad in India and stover value
                                                        chain evaluation from farmer to trader in one
Birthal, P.S., A.K. Jha, M. Tiongco, C. Delgado,
                                                        selected supply area: a rapid assessment. Report
C. Narrod, and P.K. Joshi. 2006. “Equitable
                                                        prepared for the International
Intensification of Smallholder Market-oriented
                                                        Livestock Research Institute (ILRI), Nairobi.
Dairy Production in India through Contract
                                                        Processed.
Farming.” Annex I of the IFPRI-FAO Final Report,
Contract Farming of Milk and Poultry in India:          Bosworth, B., and S. Collins. 2003. “The
Partnerships to Promote the Environmentally             Empirics of Growth: An Update.” Brookings
Friendly and Equitable Intensification                  Paper on Economic Activity. No. 2. 113–206.
of Smallholder Livestock Production.”
                                                        Bosworth, B., and S. Collins. 2008. “Accounting
International food Policy Research Institute
                                                        for Growth: Comparing China and India.”
(IFPRI), Washington, DC.
                                                        Journal of Economic Perspectives 22(1): 45–66.
Birthal, P.S., P.K. Joshi, D.S. Negi, and S. Agarwal.
                                                        Bosworth B., S. Collins, and A. Virmani. 2007.
2014. “Changing Sources of Growth In Indian
                                                        “Sources of Growth in the Indian Economy”.
Agriculture.” IFPRI Discussion Paper 01325.
                                                        NBER Working Paper No. 12901. National
International Food Policy Research Institute
                                                        Bureau of Economic Research (NBER),
(IFPRI), New Delhi.
                                                        Cambridge, MA. http://www.nber.org/papers/
Birthal, P.S., P.K. Joshi, D. Roy, and A. Thorat.       w12901.
2006. “Diversification in Indian Agriculture
                                                        Cassman, K.G. 2011. “Feed the Future: Framing
towards High Value Products: What Determines
                                                        the Issues on Spaceship Earth”. Presented at
It and Who Participates in It?” International
                                                        the Feed the Future Research Forum: Engaging
Food Policy Research Institute (IFPRI),
                                                        the Research Community, Washington, DC,
Washington, DC.
                                                        June 21–23.




  176
                                                                                             References




Cassman, K.G. and P.L. Pingali. 1995.               Chand, R., and P. Kumar. 2004. “Determinants
“Extrapolating Trends from Long-Term                of Capital Formation and Agriculture Growth:
Experiments to Farmers’ Fields: The Case of         Some New Explorations.” Economic and Political
Irrigated Rice Systems in Asia.” In V. Barnett,     Weekly 39(52): 5611–16.
R. Payne and R. Steiner (eds.), Agricultural
                                                    Chand, R., P. Kumar, and S. Kumar. 2011.
Sustainability in Economic, Environmental and
                                                    “Total Factor Productivity and Contribution of
Statistical Terms. John Wiley and Sons, Ltd.,
                                                    Research Investment to Agricultural Growth
London.
                                                    in India.” NCAP Policy Paper No. 25. National
Caves, D.W., L.R. Christensen and W.E. Diewert.     Centre for Agricultural Economics and Policy
1982. “The Economic Theory of Index Numbers         Research (NCAP), New Delhi.
and the Measurement of Input, Output and            Chand, R., and S. Parappurathu. 2011.
Productivity.” Econometrica. 50: 1393–414.          “Historical and Spatial Trends in Agriculture:
Chand, R. (2000) “Emerging Trends and Regional      Growth Analysis at National and State Level
Variations in Agricultural Investments and          in India.” IGIDR Proceedings/Projects Series
their Implications for growth and equity”,Policy    No. PP-069-3b. Indira Gandhi Institute of
paper, 11, NCAP, New Delhi.                         Development Research (IGIDR), Mumbai.

Chand, R. 2003a. “Government Intervention           Chand, R., P.A.L. Prasanna, and A. Singh. 2011.
in Food Grains Markets in India in a Changing       “Farm Size and Productivity: Understanding
Context.” Policy Paper No. 19. National Center      the Strengths of Smallholders and Improving
for Economics and Policy Research (NCAP),           Their Livelihoods.” Economic and Political Weekly.
New Delhi.                                          Supplement. 47(26&27): 5–11.

__________. 2003b. “Minimum Support Price in        Chand, R., S.S. Raju, and L.M. Pande. 2007.
Agriculture: Changing Requirements.” Economic       “Growth Crisis in Agriculture: Severity and
and Political Weekly 38(29): 3027–28.               Options at the National and State Levels.”
                                                    Economic and Political Weekly 42(26): 2528–33.
__________. 2012. “Development Policies and
Agricultural Markets.” Economic and Political       Chawla, N.K., M.G.P. Kurup, and V.P. Sharma.
Weekly 47(52), 53–63.                               2004. Animal Husbandry: State of the Indian Farmer.
                                                    A Millennium Study, Vol. 12. New Delhi:
Chand, R., and S. Bajar. 2012. “Agricultural        Academic Foundation.
Trade Liberalization Policies in India: Balancing
                                                    Chenery, H., M.S. Ahluwalia, C.L.G. Bell, J.H.
Producer and Consumer Interests.” In Twenty
                                                    Duloy, and R. Jolly. 1974. Redistribution with
Years of India’s Liberalization: Experiences and
                                                    Growth. Oxford: Oxford University Press.
Lessons, edited by R. Banga and A. Das. Geneva:
United Nations Conference on Trade and              Chenery, H.B., and M. Syrquin. 1975. Patterns
Development (UNCTAD). PP. 27–42.                    of Development, 1950–1970. London: Oxford
                                                    University Press.
Chand, R., A. Gulati, P. Shinoj, and K. Ganguly.
2010. “Managing Food Inflation in India:            Chenery, H.B., and L. Taylor. 1968.
Reforms and Policy Options.” Policy Brief No. 35.   “Development Patterns among Countries and
National Centre for Agricultural Economics and      over Time.” Review of Economics and Statistics 50(3):
Policy Research (NCAP), New Delhi.                  391–416.



                                                                                                     177
Republic of India: Accelerating Agricultural Productivity Growth




Coelli, T.J., and D.S. Prasada Rao. 2005. "Total           Datt, G., and M. Ravallion. 2009. “Has India’s
Factor Productivity Growth in Agriculture:                 Economic Growth Become More Pro-Poor
A Malmquist Index Analysis of 93 Countries,                in the Wake of Economic Reforms?” Policy
1980–2000." Agricultural Economics 32                      Research Working Paper No. 5103. World Bank,
(Supplement): 115–34.                                      Washington, DC.
Contini, E., J.G. Gasques, E. Alves, and E. Teles          Deaton, A. 1989. “Rice Prices and Income
Bastos. 2010. “Dinamismo da Agricultura                    Distribution in Thailand: A Non-Parametric
Brasileira.” Revista de Política Agrícola XIX. Brasília:   Analysis.” Economic Journal. 99:1-37.
Secretaria Nacional de Política Agrícola,
                                                           Deaton, A., and J. Dreze. 2009. “Food and
Companhia Nacional de Abastecimento.
                                                           Nutrition in India: Facts and Interpretations.”
PP.. 42–64.
                                                           Economic and Political Weekly 44(7): 42–65.
Das Gupta, S., T. Reardon, B. Minten, and
                                                                  2010. "Nutrition, Poverty, and Calorie
                                                           __________.
S. Singh. 2010a. “The Transforming Potato
                                                           Fundamentalism: Response to Utsa Patnaik."
Value Chain in India: Potato Pathways from
                                                           Economic and Political Weekly 45(14): 78–80.
a Commercialized-Agriculture Zone (Agra) to
Delhi.” Chapter 2 of “Improved Value Chains to             Dev, S.M. 2008a. Inclusive Growth in India:
Ensure Food Security in South and Southeast                Agriculture, Poverty and Human Development. New
Asia,” Asian Development Bank (ADB) and                    Delhi: Oxford University Press.
International Food Policy Research Institute
                                                                  2008b. “Challenges for Revival of Indian
                                                           __________.
(IFPRI), New Delhi.
                                                           Agriculture.” First Dr. Dayanath Jha Memorial
__________. 2010b. “The Transforming Rice                  Lecture. National Centre for Agricultural
Value Chain in India: The Rice Road from a                 Economics and Policy Research (NCAP). July.
Commercialized-Agriculture Zone in Western
                                                                  2013a. Synthesis Report on Background
                                                           __________.
Uttar Pradesh to Delhi.” Chapter 3 of “Improved
                                                           Papers for the FAO “TCP/IND/3301—Policy
Value Chains to Ensure Food Security in South
                                                           Options and Investment Priorities for
and Southeast Asia,” Asian Development Bank
                                                           Accelerating Agricultural Growth in India.”
(ADB) and International Food Policy Research
                                                           Processed.
Institute (IFPRI), New Delhi.
                                                                  2013b. “Post-2015 Development
                                                           __________.
Dasgupta, D., R.N. Dubey, and R. Sathish. 2011.
                                                           Agenda: Employment and Growth with Special
“Domestic Wheat Price Formation and Food
                                                           Reference to India.” IDS Bulletin 44(5–6): 63–71.
Inflation in India.” Working Paper No. 2/2011-
DEA. Department of Economic Affairs, Ministry              Diewert, W.E. 1976. Exact and Superlative Index
of Finance, Government of India, New Delhi.                Numbers. Journal of Econometrics 4: 115–45.
Dar, M.H., A. d Janvry, K. Emerick, D. Raitzer,                   1978. Superlative Index Numbers and
                                                           __________.

and E. Sadoulet. 2012. Flood-tolerant rice                 Consistency in Aggregation. Econometrica 46:
expected to decrease yield variability, especially         883–900.
for socially disadvantaged groups in India.
                                                           Dikshit, A.K. and P.S. Birthal. 2010. “India’s
International Rice Research Institute (IRRI),
                                                           Livestock Feed Demand: Estimates and
Manila. Processed.




  178
                                                                                              References




Projections.” Agricultural Economics and Research    Fan, S., P.B.R. Hazell, and S. Thorat. 1999.
Review. 23(1): 15-28.                                “Linkages between Government Spending,
                                                     Growth, and Poverty in Rural India.” Research
Ehui, S.K., and D.S.C. Spencer. 1993. “Measuring
                                                     Report No. 110. International Food Policy
the Sustainability and Economic Viability of
                                                     Research Institute (IFPRI), Washington, DC.
Tropical Farming Systems: A Model From Sub-
Saharan Africa.” Agricultural Economics 9: 279–96.   Färe, R., S. Grosskopf, M. Norris, and Z. Zhang.
                                                     1994. “Productivity Growth, Technical Progress,
Evenson, R.E., and K.O. Fuglie. 2010.                and Efficiency Change in Industrialized
“Technology Capital: The Price of Admission to       Countries.” American Economic Review 84(1):
the Growth Club.” Journal of Productivity Analysis   66–83.
33: 173–90.
                                                     Farrell, M.J. 1957. “The Measurement of
Evenson, R.E., and D. Jha. 1973. “The                Productive Efficiency.” Journal of the Royal
Contribution of the Agricultural Research            Statistical Society Series A. 120:253-90.
System to Agricultural Production in India.”
Indian Journal of Agricultural Economics 28(4):      FAO (Food and Agriculture Organization). 2006.
212–30.                                              Livestock's Long Shadow: Environmental Issues and
                                                     Options. Rome.
Evenson, R.E., C. Pray, and M.W. Rosegrant.
                                                            2011. World Livestock 2011: Livestock in
                                                     __________.
1999. “Agricultural Research and Productivity
                                                     Food Security. Rome.
Growth in India.” Research Report No 109.
International Food Policy Research Institute         Ferroni, M. 2013. Transforming Indian Agriculture:
(IFPRI), Washington, DC.                             India 2040. New Delhi: Sage Publications.
Fafchamps, M., R.V. Vargas-Hill, and B. Minten.      Ferroni, M., and Y. Zhou. 2012. “Achievements
2008. “Quality Control in Non-staple Food            and Challenges in Agricultural Extension in
Markets: Evidence from India.” Agricultural          India.” Global Journal of Emerging Market
Economics 38(3): 251–66.                             Economies 4(3): 319–46. 

Fairoze, M.N., L. Achoth, K.K. Prasanna Rashmi,      FICCI (Federation of Indian Chambers of
M. Tiongco, C. Delgado, C. Narrod, and C.G.          Commerce and Industry). 2010. “Corporate
Chengappa. 2006. “Equitable Intensification of       Interventions in Indian Agriculture: Towards a
Smallholder Market-oriented Poultry Production       Resilient Farming Community.” New Delhi.
in India through Contract Farming.” Annex I of       Fischer, R.A., D. Byerlee and G. O. Edmeades.
the IFPRI-FAO Final Report, “Contract Farming        2009. “Can Technology Deliver on the Yield
of Milk and Poultry in India: Partnerships           Challenge to 2050?” Paper prepared for the
to Promote the Environmentally Friendly              Expert Meeting on How to Feed the World in
and Equitable Intensification of Smallholder         2050, Food and Agriculture Organization of the
Livestock Production.” International Food Policy     United Nations, Rome, June 24-26.
Research Institute (IFPRI), Washington, DC.
                                                     Fischer, R.A., D. Byerlee, and G. Edmeades.
Fan, S., A. Gulati, and S. Thorat. 2008.             2013.“Crop Yields and Global Food Security:
“Investment, Subsidies and Pro-poor Growth in        Can Continuing Crop Yield Increases Feed the
Rural India.” Agricultural Economics. 39: 163–70.    World?” Forthcoming. ACIAR, Canberra.



                                                                                                    179
Republic of India: Accelerating Agricultural Productivity Growth




Foster, A.D., and M.R. Rosenzweig. 2010. “Is           Ganesh-Kumar, A. and K. Parikh. 1998. “A Stock-
There Surplus Labor in Rural India?” Discussion        Trade Policy for National Level Food Security
Paper No. 991. Economic Growth Center, Yale            for India.” Mimeo, Indira Gandhi Institute for
University, New Haven.                                 Development Research, Mumbai.
Freedman, D.A. 2001. “Ecological Inference and         Garg, M.R., S.A Biradar, and R. Kannan.
the Ecological Fallacy.” Vol. 6 of International       2009. “Assessment of Economic Impact of
Encyclopedia of the Social and Behavioral Sciences,    Implementing Ration Balancing Program
edited by N.J. Smelser and P.B. Baltes.                on Lactating Cattle and Buffalo Under Field
PP. 4027–30.                                           Conditions.” Indian Journal Nutrition 26(2):
                                                       146–150.
Freedman, D., R. Pisani, and R. Purves. 2007.
Statistics. 4th ed. New York: W.W. Norton.             Gaurav, S. and S. Mishra. 2011. “Size-Class and
                                                       Returns to Cultivation in India: A Cold Case
Fuglie, K.O. 2012. “Productivity Growth in
                                                       Reopened.” Working Paper 27, Indira Gandhi
the Global Agricultural Economy and the
                                                       Institute of Development Research, Mumbai.
Role of Technology Capital.” In Productivity
Growth in Agriculture: An International Perspective,   Gouel, C., M. Gautam and W. J. Martin. 2014.
edited by K.O. Fuglie, S.L. Wang, and V.E. Ball.       “India – Managing Wheat Price Policies:
Wallingford, UK: CAB International.                    Evaluating Policy Options.” Mimeo. Agriculture,
Gaiha, R., N. Kaicker, K.S. Imai, and G. Thapa.        Irrigation and Natural Resources Unit,
2012. “Demand for Nutrients in India: An               South Asia Region, The World Bank,
Analysis Based on the 50th, 61st, and 66th             Washington, DC.
Rounds of the NSS.” Discussion Paper No. 1211.         GOI (Government of India). 1999. “Common
Economics, School of Social Sciences, University       Property Resources in India.” Report No.
of Manchester.                                         452(54/31/4). National Sample Survey
Gandhi, V.P., and Z.-Y. Zhou. 2010. “Rising            Organization, Ministry of Statistics and
Demand for Livestock Products in India:                Programme Implementation, New Delhi.
Nature, Patterns, and Implications.” Australian               2006a. “Regional Disparities and Rural
                                                       __________.

Agribusiness Review 18: 103–35.                        Distress in Maharashtra.” Report of the Planning
Ganesh-Kumar, A., A. Gulati, and R. Cummings,          Commission’s Fact Finding Team on Vidharbha,
Jr. 2007. “Foodgrains Policy and Management            Maharashtra.
in India: Responding to Today’s Challenges                    2007. “Report of the Steering Committee
                                                       __________.

and Opportunities.” International Food Policy          on Agriculture and Allied Sectors for the
Research Institute (IFPRI), Washington, DC.            Formulation of the Eleventh Fiver Year Plan
Ganesh-Kumar, A., R. Mehta, H. Pullabhotla,            2007–2012.” Planning Commission, New Delhi.
S.K. Prasad, K. Ganguly, and A. Gulati. 2012.                  2008a. “Eleventh Five Year Plan (2007-
                                                       __________.

“Demand and supply of cereals in India: 2010-          2012).” Planning Commission, New Delhi.
2025.” Discussion Paper No. 1158. International
Food Policy Research Institute (IFPRI),                        2008b. “Basic Animal Husbandry
                                                       __________.


Washington, DC.                                        Statistics.” Government of India, New Delhi.




  180
                                                                                              References




       2011a. “Mid-Term Appraisal of the
__________.
                                                    Gulati, A., K. Ganguly, and G. Shreedhar.
Eleventh Five Year Plan.” Planning Commission,      2011. “Food and Nutritional Security in India:
New Delhi.                                          A Stocktaking Exercise.” Report submitted
                                                    to ICAR. International Food Policy Research
        2011b. “Faster, Sustainable and More
__________.
                                                    Institute, New Delhi.
Inclusive Growth: An Approach to the Twelfth
Five Year Plan.” Planning Commission,               Gulati, A., S. Jain, and A. Hoda. 2013. “Farm
New Delhi.                                          Trade: Tapping the Hidden Potential.”
                                                    Discussion Paper No. 3. Commission for
       2011c. “Report of the Working Group on
__________.
                                                    Agricultural Costs and Prices (CACP), Ministry of
Agricultural Research and Education.” For the
                                                    Agriculture, Government of India, New Delhi.
Twelfth Five Year Plan. Planning Commission,
New Delhi.                                          Gulati, A., S. Jain, and N. Saltija. 2013. “The
                                                    Rising Farm Wages in India: The ‘Pull’ and
       2012. “Twelfth Five Year Plan
__________.
                                                    ‘Push’ Factors.” Discussion Paper No. 5.
(2012–17) – Volume 2 (Economic Sectors).”
                                                    Commission for Agricultural Costs and Prices
Planning Commission, New Delhi.
                                                    (CACP), Ministry of Agriculture, Government of
       2013a. “Economic Survey.” Ministry of
__________.
                                                    India, New Delhi.
Finance, New Delhi.
                                                    Gulati, A., M. Landes, and K. Ganguly. 2009.
       2013b. “Price Policy for Kharif Crops:
__________.
                                                    "Indian Agriculture: Managing Growth with
Marketing Season 2013–14.” Commission for           Equity." Choices. 2nd Quarter.
Agricultural Costs and Prices (CACP), Ministry of
                                                    Gulati, A., and S. Narayanan. 2003. Subsidy
Agriculture, New Delhi.
                                                    Syndrome in Indian Agriculture. New Delhi: Oxford
Grace, D., D. Baker, and T. Randolph.               University Press.
2010. “Innovative and Participatory Risk-
                                                    Haileslassie, A., M. Blümmel, F. Clement,
based Approaches to Assess Milk Safety
                                                    S. Ishaq, and M.A. Khan. 2011. “Adapting
in Developing Countries: A Case Study in
                                                    Livestock Water Productivity to Climate
Northeast India.” In Demand for Livestock
                                                    Change.” International Journal of Climate Change
Products in Developing Countries with a Focus on
                                                    Strategies and Management 3(2): 156–69.
Quality and Safety Attributes: Evidence from Asia
and Africa, edited by M.A. Jabbar, D. Baker,        Haque, T. 2012. Land Policies for Inclusive Growth.
and M.L. Fadiga. ILRI Research Report No. 24.       New Delhi: Concept Publishing.
Nairobi: International Livestock Research
                                                            2013. “Impact of Land Leasing
                                                    __________.
Centre (ILRI). PP. 116-26.
                                                    Restrictions on Agricultural Efficiency and
Griliches, Z. 1964. “Research Expenditures,         Equity in India.” Paper presented at the Annual
Education, and the Aggregate Agricultural           World Bank Conference on Land and Poverty,
Production Function.” American Economic Review      Washington, DC, April l8–11. Also available at:
54: 961–74.                                         http://www.landandpoverty.com/agenda/pdfs/
                                                    paper/haque_full_paper.pdf.
Gulati, A., and K. Ganguly. 2010. “The Changing
Landscape of Indian Agriculture.” Agricultural      Harper, L.A., O.T. Denmead, J.R. Freney, and
Economics 41(777): 37–45.                           F.M. Byers. 1999. “Direct Measurements of



                                                                                                    181
Republic of India: Accelerating Agricultural Productivity Growth




Methane Emissions from Grazing and Feedlot          Countries.” Asian Journal of Agriculture and
Cattle.” Journal of Animal Science 77: 1392–1401.   Development. 1(1).
Hazell, P.B.R., D. Headey, A. Nin-Pratt, and        ICRISAT (International Crops Research Institute
D. Byerlee. 2011. Structural Imbalances and         for the Semi-Arid Tropics). 2005. “Overcoming
Farm and Non-Farm Employment Prospects in           Poverty in Rural India: Focus on Rainfed Semi-
Rural South Asia. Background paper for a World      Arid Tropics.” Patancheru.
Bank South Asia Region flagship report, “Where
                                                            2013. “District Level Database
                                                    __________.
Are the More and Better Jobs?” Processed.
                                                    Documentation.” Village Dynamics in South
Herrero M., P.K. Thornton, A. Notenbaert,           Asia, ICRISAT-ICAR-IRRI Collaborative Research
S. Msangi, S. Wood, R.L. Kruska, J. Dixon, D.       Project. Patancheru.
Bossio, J.A. van de Steeg, H.A. Freeman, X. Li,
                                                    Imai, K., R. Gaiha, and G. Thapa. 2008.
and P.P. Rao. 2009. “Drivers of Change in Crop-
                                                    “Transmission of World Commodity Prices
Livestock Systems and Their Potential Impacts
                                                    to Domestic Commodity Prices in India and
on Agro-ecosystems Services and Human Well-
                                                    China.” BWPI Working Paper No. 45. Brooks
being to 2030: A Study Commissioned by the
                                                    World Poverty Institute (BWPI), University of
CGIAR Systemwide Livestock Program.” ILRI
                                                    Manchester.
Project Report. International Livestock Research
Institute (ILRI), Nairobi.                          Indian Immunologicals. 2012. Annual Report.
                                                    Hyderabad.
Himanshu, P. Lanjouw, A. Mukhopadhyay, and
R. Murgai. 2010: “Non-farm Diversification and      IPCC (Intergovernmental Panel on Climate
Rural Poverty Decline: A Perspecive from Indian     Change). 2013. “Climate Change 2013: The
Sample Survey and Village Study Data.” Asia         Physical Science Basis.” Contribution of
Research Centre Working Paper 44. London            Working Group I to the Fifth Assessment Report
School of Economics and Political Science,          of the Intergovernmental Panel on Climate
London.                                             Change. Cambridge: Cambridge University
                                                    Press.
HLC (High Level Committee on Estimation
of Saving and Investment). 2009: “Report of         Ivanic, M. and W. Martin. 2008. “Implications
High Level Committee on Estimation of Saving        of Higher Global Food Prices for Poverty in
and Investment.” Ministry of Statistics and         Low-Income Countries.” Agricultural Economics.
Programme Implementation, Government of             39:405-16.
India, New Delhi.
                                                    Ivanic, M., W. Martin and H. Zaman. 2011.
Hoda, A., and A. Gulati. 2013. “India’s             “Estimating the Short-Run Poverty Impacts
Agricultural Trade Policy and Sustainable           of the 2010-11 Survey in Food Prices.” Policy
Development Goals.” Issue Paper No. 48.             Research Working Paper No. 5633, The World
International Centre for Trade and Sustainable      Bank, Washington, DC.
Development, Geneva.
                                                    Jacoby, H. 2013. “Food Prices, Wages and
Huang, J., R. Hu and S. Rozelle. 2004. “China’s     Welfare in Rural India.” Policy Research
Agricultural Research System and Reforms:           Working Paper Bo. 6412. Development Research
Challenges and Implications for the Developing      Group, World Bank, Washington, DC.



  182
                                                                                                  References




Jha, D. and S. Kumar (2006), Research Resource           for Food Commodities in India.” Agricultural
Allocation in Indian Agriculture, National               Economics Research Review 24(1): 1–14.
Centre for Agricultural Economics and Policy
                                                         Kumar, P. and S. Mittal. 2006. “Agriculture
Research, NCAP Policy Paper 23, New Delhi.
                                                         Productivity Trends in India: Sustainability Issues.”
Jha, S., and P.V. Srinivasan. 1999. “Grain Price         Agricultural Economics Research Review. 19: 71–88.
Stabilization in India: Evaluation of Policy
                                                         Kurukulasuriya, P., and S. Rosenthal. 2003.
Alternatives.” Agricultural Economics, 21(1):
                                                         “Climate Change and Agriculture: A Review of
93–108.
                                                         Impacts and Adaptations.” World Bank Change
Jin, S., H. Hengyun Ma, J. Huang, R. Hu, and             Series Paper No. 91. Rural Development Group
S. Rozelle. 2010. “Productivity, Efficiency and          and Environment Department, World Bank,
Technical Change: Measuring the Performance              Washington, DC.
of China’s Transforming Agriculture”. Journal of
                                                         Kuznets, S. 1955. “Economic Growth and Income
Productivity Analysis 33:191–207.
                                                         Inequality.” American Economic Review 49 (1): 1–28.
Johnston, B.F., and J.W. Mellor. 1961. “The Role
                                                               1966. Modern Economic Growth.
                                                         __________.
of Agriculture in Economic Development.”
                                                         New Haven: Yale University Press.
American Economic Review.
                                                         Landes, M.R., and M.E. Burfisher. 2009.
Joshi, P.K., and A. Gulati. 2007. "From Plate to
                                                         “Growth and Equity Effects of Agricultural
Plow: Agricultural Diversification in India." In
                                                         Marketing Efficiency Gains in India.” USDA-ERS
The Dragon and The Elephant: Agricultural and Rural
                                                         Economic Research Report No. 89. United States
Reforms in China and India, edited by A. Gulati and
                                                         Department of Agriculture (USDA), Economic
S. Fan. Baltimore: Johns Hopkins.
                                                         Research Service (ERS), Washington, DC.
Joshi, P.K., A. Gulati, and R. Cummings, Jr. 2007.
                                                         Lanjouw, P., and R. Murgai. 2009. “Poverty
Agricultural Diversification and Smallholders in South
                                                         Decline, Agricultural Wages, and Non-Farm
Asia. New Delhi: Academic Foundation.
                                                         Employment in Rural India: 1983–2004.”
Joshi, P.K., and P. Kumar. 2011. Food demand             Agricultural Economics. 40: 243–263.
and supply projections for India. International
                                                         Lele, U., and A.A. Goldsmith. 1989. "The
Food Policy Research Institute (IFPRI),
                                                         Development of National Agricultural
New Delhi. Processed.
                                                         Research Capacity: India's Experience with the
Kampman, D.A. 2007. The Water Footprint of               Rockefeller Foundation and Its Significance for
India. Master Thesis. Department of Water                Africa." Economic Development and Cultural Change
Engineering and Management, University of                37(2): 305–43.
Twente, Enschede, The Netherlands.
                                                         Lele, U., and J.W. Mellor. 1981. “Technological
Kotwal, A., B. Ramaswamy, and W. Wadhwa.                 Change, Distribution Bias and Labour Transfer
2011. “Economic Liberalization and Indian                in a Two-Sector Economy.” Oxford Economic Papers
Economic Growth: What is the Evidence?”                  33: 426–41.
Journal of Economic Literature. 49(4): 1152–99.
                                                         Lewis, W.A. 1954. “Economic Development with
Kumar, P., A. Kumar, S. Parappurathu, and S.S.           Unlimited Supplies of Labor.” Manchester School
Raju. 2011. “Estimation of Demand Elasticity             22(2): 139–91.



                                                                                                        183
Republic of India: Accelerating Agricultural Productivity Growth




Lobell, D.B., K.G. Cassman and C.R. Field. 2009.     Morisset, M., and P. Kumar. 2008a. Structure and
“Crop Yield Gaps: Their Importance, Magnitudes       performance of the food processing industry in
and Causes.” NCESR Publications and Research.        India. Processed.
Paper 3. http://digitalcommons.unl.edu/
                                                            2008b. Trends and pattern of
                                                     __________.
ncesrpub/3.
                                                     consumption of value added foods in India.
Makkar, H.P.S. 2012. “Quest for Novel                Processed.
Livestock Feeds.” Plenary presentation for the
                                                     Murgai, R., M. Ali, and D. Byerlee. 2001. “Sense
Technical Consultation Meeting on Assessing
                                                     and Sustainability in Post-Green Revolution
the Potential of Insects as Food/Feed in
                                                     Agriculture: The Case of Indian and Pakistan
Assuring Food Security, Food and Agriculture
                                                     Punjabs.” World Bank Research Observer. 16(2):
Organization of the United Nations (FAO),            199–218.
Rome, January 23–25.
                                                     NAAS (National Academy of Agricultural
Mattoo, A., D. Mishra, and A. Narain. 2007.          Sciences). 2010. “State of Indian Agriculture I.”
From Competition at Home to Competing Abroad: A      New Delhi.
Case Study of India’s Horticulture. Oxford: Oxford
University Press.                                          2011. “State of Indian Agriculture II.”
                                                     __________.

                                                     New Delhi.
Mellor, J.W., and U. Lele. 1973. “Growth
Linkages of New Food Grain Technologies.”            NAAS (National Academy of Agricultural
Indian Journal of Agricultural Economics 28(1):      Sciences) and ICAR (Indian Council for
35–55.                                               Agricultural Research). 2013. “Bhubaneswar
                                                     Declaration.” New Delhi
Minot, N., M. Epprecht, T. T. Tram Anh and L. Q.
Trung. 2006. Income diversification and poverty      Narain, U., S. Gupta, and K. Veld. 2005.
in Northern Uplands of Vietnam. Research             “Poverty and the Environment: Exploring the
Report 145. International Food Policy Research       Relationship between Household Incomes,
Institute, Washington DC.                            Private Assets, and Natural Assets.” Discussion
                                                     Paper No. 05-18. Resources for the Future,
Minten, B. 2008. “The Food Retail Revolution         Washington, DC.
in Poor Countries: Is It Coming or Is It Over?”
Economic Development and Cultural Change 56(4):      Nene, Y.L., and P.M. Tamboli. 2011. "Revitalizing
767–89.                                              Higher Agricultural Education in India: Journey
                                                     towards Excellence." Asian Agri-History
Minten, B., T. Reardon, and R. Sutradhar. 2010.      Foundation, Hyderabad.
“Food Prices and Modern Retail: The Case of
Delhi.” World Development 38(12): 1775–87.           Nimbkar, C., and N. Kandasamy. 2012. “Animal
                                                     Breeding in India: A Time for Reflection, and
Mishra, S. 2008. “Risks, Farmer Suicides, and        Action.” Journal of Animal Breeding and Genetics
Agrarian Crisis in India: Is There a Way Out?”       128(3): 161–62.
Indian Journal of Agricultural Economics. 63(1):
                                                     Nin-Pratt, A., B. Yu and S. Fan. 2009.
38–54.
                                                     “Comparisons of Agricultural Productivity
Mohanty, D. 2011. “Infaltion: Decoding the           Growth in China and India.” Journal of
Dynamics.” RBI Monthly Bulletin November.            Productivity Analysis 33(3): 209–23.



  184
                                                                                             References




Palaskas, T.B., and B. Harriss-White. 1996. “The     Association and Canadian Agricultural
Identification of Market Exogeneity and Market       Economics Society Joint Annual Meeting,
Dominance by Tests instead of Assumption:            Washington, DC, August 4–6.
An Application to Indian Material.” Journal of
                                                     Rada, N., and K. Fuglie. 2012. “Shifting Sources
International Development 8(1): 11–23.
                                                     of Agricultural Growth in Indonesia: A Regional
Parikh, K., H. Binswanger-Mkhize, P. Ghosh, and      Analysis.” Chapter 10 in Productivity Growth in
A. D’Souza. 2011. “Structural Transformation         Agriculture: An International Perspective, edited by
of the Indian Economy and Its Agriculture:           K.O. Fuglie, S.L. Wang, and V.E. Ball. Wallingford,
1961–2039.” Paper prepared for the Centennial        UK: CAB International. PP. 199–214.
Group, Integrated Research and Action for
                                                     Rada, N., and C. Valdes. 2012. “Policy,
Development (IRADe), Mumbai.
                                                     Technology, and Efficiency of Brazilian
Peden, D., G. Tadesse, and A.K. Misra.               Agriculture.” Economic Research Report
2007. “Water and Livestock for Human                 No. 137. United States Department of
Development.” Chapter 13 in Water for Food,          Agriculture (USDA), Economic Research Service
Water for Life: A Comprehensive Assessment of        (ERS), Washington, DC.
Water Management in Agriculture, edited by D.
Molden. London and Colombo: Earthscan and            Rajmal, and S. Misra. 2009. “Transmission from
International Water Management Institute             International Food Prices to Domestic Food
(IWMI). PP 485–514.                                  Prices – The Indian Evidence.” RBI Staff Studies.
                                                     Department of Economic Analysis and Policy,
Pray, K., and L. Nagarajan. 2012. “Innovation        Reserve Bank of India (RBI), New Delhi.
and Research by Private Agribusiness in India.”
IFPRI Discussion Paper No. 1181. International       Ramachandra, K.V., V.K. Taneja, K.T. Sampath,
Food Policy Research Institute (IFPRI),              S. Anandan, and U.B. Angadi. 2007. Livestock
Washington, DC.                                      feed resources in different agro-ecosystems
                                                     of India: Availability, requirement and their
Pursell, G., A. Gulati, and K. Gupta. 2009.          management. National Institute of Animal
“Distortions to Agricultural Incentives in India.”   Nutrition and Physiology, Bangalore. Processed.
In Distortions to Agricultural Incentives in Asia,
edited by K. Anderson and W. Martin. World           Ramasamy, C. 2012. “Indian Agricultural
Bank: Washington, DC. PP. 339–78.                    R&D: An Instrospection and Way Forward.”
                                                     Presidential Address, 20th Annual Conference of
Raabe, K. 2008. “Reforming the Agricultural          Agricultural Economics Research Association,
Extension System in India: What Do We
                                                     held at IARI, New Delhi, October 9.
Know about What Works, Where, and Why?”
Discussion Paper No. 00775. International            Ramaswami, B., and P. Balakrishnan. 2002.
Food Policy Research Institute (IFPRI),              “Food Prices and the Efficiency of Public
Washington, DC.                                      Intervention: The Case of Public Distribution
                                                     System in India.” Food Policy 27(5–6): 419–36.
Rada, N. 2013. “Agricultural Growth in
India: Examining the Post-Green Revolution           Rao, N.C., and S.M. Dev. 2010. “Agricultural Price
Transition.” Selected Paper prepared for the         Policy, Farm Profitability and Food Security.”
2013 Agricultural and Applied Economics              Economic and Political Weekly 45, June 26.



                                                                                                   185
Republic of India: Accelerating Agricultural Productivity Growth




Rashid, S., A. Gulati, and R. Cummings. 2008.        Growth in South Asia.” American Journal of
From Parastatals to Private Trade: Lessons from      Agricultural Economics 74 (3): 757–61.
Asian Agriculture. Baltimore: Johns Hopkins.
                                                              1995. “Total Factor Productivity and
                                                     __________.

Raturi, R. 2011. “Implementation Review of           Sources of Long-term Growth in Indian
Major Centrally Funded Programs.” Background         Agriculture.” Discussion Paper No. 7.
paper for Syngenta Foundation project on             International Food Policy Research Institute
the “Future of Indian Agriculture and Poverty        (IFPRI), Washington, DC.
Reduction.” New Delhi.
                                                     SAPPLPP (South Asia Pro-Poor Livestock Policy
Ray D.K., N.D. Mueller, P.C. West, and J.A. Foley.   Programme). 2011. “Common Land and Poor
2013 “Yield Trends Are Insufficient to Double        Livestock Keepers Experiences from Common
Global Crop Production by 2050.” PLoS ONE            Land Development in the States of Rajasthan
8(6):e66428. doi: 10.1371/journal.pone.0066428.      and Madhya Pradesh in India.” Document 021.
Reardon, T. and B. Minten. 2013. “The Quiet          New Delhi.
Revolution in India’s Food Supply Chains.”                 2012a. “Small Ruminant Rearing: Breed
                                                     __________.

In Ferroni, M. (ed). 2013, Transforming Indian       Conservation and Genetic Improvement.”
Agriculture: India 2040. New Delhi: Sage             New Delhi.
Publications.
                                                            2012b. “Small Ruminant Rearing Product
                                                     __________.

Reardon, T., P. Timmer, and B. Minten. 2010.         Markets, Opportunities and Constraints.” New
“Supermarket Revolution in Asia and Emerging         Delhi.
Development Strategies to Include Small
Farmers.” Proceedings of the National Academy        Schultz, T.W. 1953. Economic Organization of
of Sciences 109(31): 12332–37.                       Agriculture, New York: McGraw-Hill.

Rawal, V. 2008. “Ownership Holdings of Land          Sekhar, CSC. 2012. “Agricultural Market
in Rural India: Putting the Record Straight.”        Integration in India: An Analysis of Select
Economic and Political Weekly. March.                Commodities.“ Food Policy. 37: 309–22.

Robinson, T., and H.P.S. Makkar. 2012. “Demand       Sekhri., S. 2013. Missing water: Agricultural
Growth for Animal-source Foods: Implications for     stress and adaptation strategies in response to
Livestock Feed Production. In Conducting National    groundwater depletion in India. Department
Feed Assessments, edited by M.B. Coughenour and      of Economics, University of Virginia,
H.P.S. Makkar. Animal Production and Health          Charlottesville. Processed.
Manual No. 15. Rome: Food and Agriculture            Sen, A. 2011. “Keynote Address,” Workshop
Organization (FAO). PP. 59–65.
                                                     on Policy Options and Investment Priorities
Rodell, M., I. Velicogna, and J.S. Famiglietti.      for Accelerating Agricultural Productivity and
2009. “Satellite-Based Estimation of                 Development in India, India International
Groundwater Depletion in India.” Nature              Center, New Delhi, November 10–11.
460(7258): 999–1002.
                                                     Shah, T. 2007. The groundwater economy of
Rosegrant, M.W., and R.E. Evenson. 1992.             South Asia: An assessment of size, significance,
“Agricultural Productivity and Sources of            and socio-ecological impacts. International



  186
                                                                                                 References




Water Management Institute, Colombo.                    services in India. Background paper prepared
Processed.                                              for the World Bank. Processed.
Shah, T., A. Gulati, P. Hemant, G. Shreedhar, and       Solow, R.M. 1975. “Technical Change and
R.C. Jain. 2009. “Secret of Gujarat’s Agrarian          Aggregate Production Function.” Review of
Miracle after 2000.” Economic and Political Weekly      Econometrics and Statistics 39(3): 312–20.
44(52): 45–55.
                                                        Srinivasan, P.V., and S. Jha. 2001. “Liberalized
Sharma, V.P. and V. Ahuja. 2005. “The Changing          Trade and Domestic Price Stability: The Case of
Structure of Dairy and Meat Value Chains in             Rice and Wheat in India.” Journal of Development
India.” Mimeo, FAO, Bangkok.                            Economics 65(2): 417–41.
Shiva, V. 2007. “Corporate Hijack of Retail:            Staal, S., A. Nin-Pratt, and M.A. Jabbar. 2008.
Retail Dictatorship vs. Retail Democracy.”              Dairy Development for the Resource Poor, Part 1: A
Navdanya and Research Foundation for Science,           Comparison of Dairy Policies in South Asia and East
Technology and Ecology, New Delhi.                      Africa, Rome and Nairobi: Food and Agriculture
Shreedhar, G., N. Gupta, H. Pullabhotla,                Organization (FAO) and International Livestock
A. Ganesh-Kumar, and A. Gulati. 2011.                   Center for Africa (ILRI).
“A Review if Input and Output Policies for              Sulaiman, V.R., and A. Hall. 2008. “The Fallacy
Cereals Production in India.” Discussion Paper          of Universal Solutions in Extension: Is ATMA the
No. 01159. International Food Policy Research
                                                        New T&V?” LINK News Bulletin, September.
Institute (IFPRI), New Delhi.
                                                        Swanson, B., K.M. Hall, and M.N. Reddy. 2008.
Singh, A. 2011. “The Changing Landscape
                                                        “A Decentralized, Participatory, Market-driven
of Public Expenditure and Investments in
                                                        Extension System: The ATMA Model in India.”
Indian Agriculture.” Indian Journal of Agricultural
                                                        Paper presented at the International Food
Economics 66(3): 301–13.
                                                        Policy Research Institute workshop, Advancing
Singh, K.M., and B. Swanson. 2006. “Developing          Agriculture in Developing Countries through
Market Driven Extension System in India.”               Knowledge and Innovation, Addis Ababa,
In Proceedings of the 22nd Annual Conference:           7–9 April.
International Teamwork in Agricultural and
Extension Education, edited by J.R. Vreyens. N.P.:      Thomas, S. 2003. Agricultural commodity
Association of International Agricultural and           markets in India: Policy issues for growth.
Extension Education (AIAEE). PP. 627–37.                Indira Gandhi Institute for Development
                                                        Research (IGIDR), Goregaon. Processed.
Singh, O.P. 2004. “Water Productivity of Milk
Production in North Gujarat, Western India.”            Timmer, C.P. 2009. A World without Agriculture:
In Proceedings of the 2nd Asia Pacific Association of   The Structural Transformation in Historical
Hydrology and Water Resources (APHW) Conference,        Perspective. Washington, DC: AEI Press.
Singapore, 5–8 July. http://rwes.dpri.kyoto-u.          Timmer, C.P., and S. Akkus, 2008. "The
ac.jp/~tanaka/APHW/APHW2004/proceedings/                Structural Transformation as a Pathway out
SUW/56-SUW-A281/56-SUW-A281.pdf.
                                                        of Poverty: Analytics, Empirics, and Politics."
Sirohi, S., A. Kumar, S. Gokhale, K. Elumalai,          Working Papers No. 150, Center for Global
G. Sinha, and I. Wright. 2008. Livestock support        Development (CGD), Washington, DC.



                                                                                                       187
Republic of India: Accelerating Agricultural Productivity Growth




Umali-Deininger, D.L., and K. Deininger. 2001.                 2007b. World Development Report 2008:
                                                       __________.

“Towards Greater Food Security for India’s Poor:       Agriculture for Development. Washington, DC.
Balancing Government Intervention and Private
                                                             2007c. “India - National Agricultural
                                                       __________.
Competition.” Agricultural Economics 25(2–3):
                                                       Technology Project.” Implementation
321–35.
                                                       Completion and Results Report. The World
Umali-Deininger, D., and M. Sur. 2007. “Food           Bank, Washington, DC.
Safety in a Globalizing World: Opportunities
                                                              2007d. “India’s Emergent Horticulture
                                                       __________.
and Challenges for India.” In Proceedings of the
                                                       Exports: Addressing Sanitary, Phyto-Sanitary
26th Conference of the International Association of
                                                       Standards and Other Challenges.” Report
Agricultural Economists, Queensland, August 12–18.
                                                       36178-IN. South Asia Agriculture and Rural
Virmani, A. 2006. “India’s Economic Growth             Development Department, The World Bank,
History: Fluctuations, Trends, Break Points and        Washington, DC.
Phases.” Indian Economic Review 41(1): 81–103.
                                                              2008a. “Demand-led Transformation of
                                                       __________.

Ward, P.S., D.L. Ortega, D.J. Spielman and             the Livestock Sector in India: Achievements,
V. Singh. 2013. “Farmer Preferences for                Opportunities and Challenges.” South Asia
Drought Tolerance in Hybrid versus Inbred              Agriculture and Rural Development Report
Rice: Evidence from Bihar, India.” Discussion          No. 48412-IN. Washington, DC.
Paper 01307, International Food Policy Research
Institute, Washington, DC.                                    2008b. “India: Taking Agriculture to the
                                                       __________.

                                                       Market.” Report No. 35953-IN. Agriculture and
WGAHD (Working Grop on Animal Husbandry                Rural Development Unit, South Asia Region.
and Dairying). 2012. Report of the Working             Washington, DC.
Group on Animal Husbandry and Dairying,
12th Five Year Plan (2012–17). Submitted to the        ______. 2009. “India: Demand Led
Planning Commission, Government of India,              Transformation of the Livestock Sector in India
New Delhi. http://planningcommission.gov.              Achievements, Challenges, and Opportunities.”
in/aboutus/committee/wrkgrp12/agri/AHD_                Agriculture and Rural Development Unit, South
REPORT_Final_rev.pdf.                                  Asia Region, Washington, DC.

World Bank. 1999. “India Foodgrain Marketing                   2010. “Perspectives on Poverty in India:
                                                       __________.


Policies: Reforming to Meet Food Security              Stylized Facts from Survey Data.” India Poverty
Needs.” Report No. 18329-IN, Rural Development         Assessment Report. Washington, DC.
Sector Unit, South Asia Region. Washington, DC.                2012a. “More and Better Jobs in South
                                                       __________.


       2003. “Improving the Investment Climate
__________.                                            Asia.” South Asia Development Matters Series.
in India.” The World Bank, Washington DC.              The World Bank, Washington, DC.

         2005. India: Re-Energizing the Agricultural
__________.                                                   2012b. “Identifying Investment
                                                       __________.


Sector to Sustain Growth and Reduce Poverty.           Opportunities for Ruminant Livestock Feeding
New Delhi: Oxford University Press.                    in Developing Countries.” Working Paper
                                                       No. 74154. Washington, DC.
       2007a. “Bihar Agriculture: Building on
__________.

Emerging Models of Success.” Working Paper             World Economic Forum. 2012. The Global
No. 36325. Washington, DC.                             Competitiveness Report 2012–2013. Geneva.



   188
annexes
                      Annex 1: Policy Reforms and
                      Private Sector Response



 Table A.1.1 Major agricultural policy reforms since 1990/91

  Year                                                     Reform
            zz The new economic policy introduces major economic reforms (including exchange
              rate deregulation, trade liberalization, export promotion, and decontrol of domestic
  1991-92
              manufacturing), with significant impact on agriculture, though the sector is not directly
              affected by the initial reforms
            zz Potassiumand phosphate fertilizer production liberalized; nitrogen (urea) pricing remains
  1992/93      unchanged due to resistance to reforms
            zz Removal of marketing restrictions in the dairy sector

            zz Agricultural trade liberalization starts after India becomes a signatory to the WTO agreement
               on agriculture and slowly removes import and export restrictions, with a notable opening up
  1994/95      of exports for common rice and subsequently wheat
            zz Trade progressively liberalized but remains heavily regulated, with frequent changes in
               response to domestic supply situation
            zz Introduction  of Targeted Public Distribution System to improve targeting and delivery of the
               Public Distribution
  1997/98
            zz FDI in wholesale trade opened up
            zz Seed Policy changes allow entry of private sector, including major international firms

  1998/99   zz Cold   Storage Order 1964 repealed
            zz De-licensingof Food Processing Industry to remove the reservation on food processing for
  1999/00
              small-scale enterprises
  2000/01   zz Quantitative   restrictions on imports eliminated
            zz Restrictions on domestic and foreign investment (FDI) (up to 100%) in bulk handling and
               storage removed
  2001/02
            zz Inter-Ministerial Task Force and Committee of State Ministers on Agricultural Marketing
               Reforms implemented
            zz Licensing requirements; stocking limits; movement restrictions on food grains and edible oils
               removed; and selective credit controls lifted
            zz Milk and Milk Products Control Order (MMPO) amendment removes restrictions on milk
  2002/03      processing capacity while continuing to regulate health and safety conditions
            zz Leather and leather and paper products removed from small-scale reservation list
            zz New Pricing Policy Schemes introduced for fertilizers
            zz New Seed Policy introduced




                                                                                                           189
Republic of India: Accelerating Agricultural Productivity Growth




Year                                                      Reform
          zz Ban on futures trading of 54 commodities, including rice, wheat, oilseeds, and pulses, removed
          zz Levy on sugar reduced from 15% to 10%
          zz Model Act for State Agriculture Produce Marketing (Development and Regulation) formulated
          zz Processed food items exempted from licensing under Industries (Development and
2003/04      Regulations) Act 1951, except those reserved for small-scale industries (SSI) and alcoholic
             beverages
          zz Food processing included in priority list for bank lending
          zz Automatic approval for 100% FDI for most processed foods, except alcohol and beer and those
             reserved for SSI
2004/05   zz Group    of Ministers established to formulate modern integrated food law
                    Horticulture Mission (NHM) initiated
          zz National
2005/06               Warehouse Receipt and Warehousing (Development and Regulation) Bill 2005
          zz Negotiable
            approved by Cabinet
          zz Food Safety and Standards Act approved
          zz CessAct repealed
2006/07   zz Forward Contracts (Regulation) Amendment Bill submitted to Parliament
          zz Futures trading prohibited for main food grains (rice, wheat, and tur and urad pulses)
          zz Export bans imposed on rice and wheat

          zz National   Food Security Mission (NFSM) launched
2007/08
          zz Rashtriya   Krishi Vikas Yojna (RKVY)
2008/09   zz Futures   trading resumed in some food commodities, including wheat
2010/11   zz Nutrient   Based fertilizer subsidies introduced
2011/12   zz Export   bans on wheat and rice lifted
2012/13   zz National   Food Security Bill (NFS) approved by Parliament


Policy Change and Private Sector                          for example the entry of PepsiCo (in 1989)
Response                                                  and McDonald’s (in 1996). These two major
                                                          investments (the former primarily foreign
The 1990s saw tremendous change in the                    equity and the latter domestic—through
Indian economy. Agricultural reforms started              franchising) generated ripple effects with
some years later in the mid-1990s, but the                the rapid emergence of snack foods (led by
broader economic reforms had already set in               chips) and fast foods through the quick service
motion important changes in the sector. Rapid             restaurant (QSR) chains. The ripples backwards
income growth provided a strong stimulus                  were through contract farming and the
for agricultural demand, and the change in                ancillary industry to supply the raw materials or
economic policies (in particular the protection           intermediate prepared food products.
of manufacturing) led to a significant reduction
in the domestic distortions in incentives for             Three key policy changes were the permission
farmers. Trade liberalization and the prospects           of FDI in wholesale trade (1997), de-licensing of
of rapid income growth led to private sector              Food Processing Industries(1999), and the New
investments in consumer products, including               Seed Policy (1998). For agriculture, these policies



 190
                                                      ﻿Annex 1: Policy Reforms and Private Sector Response




induced a number of potentially transformative         sector seeds supply is rapidly growing; it now
private sector initiatives and investments.            provides the bulk of improved seed varieties for
Several global and domestic corporate investors        horticulture and introduced the revolutionary
entered the agro-processing sector (such as            Bt Cotton (in 2002) that has since transformed
PepsiCo, Coca-Cola, Cargill, Hindustan Unilever,       the Indian cotton sector.
Reliance, Tata, Godrej, ITC, and Mahindra’s).
McCain, the world’s largest producer of French         Recognizing the importance of key services—
fries and potato specialties, started operations       ostensibly to fill the large void in public service
in 1998 with a focus on the frozen food market         delivery in agricultural market information
for India and the Indian subcontinent.                 and extension (Ahuja 2014; Ferroni 2013)—
                                                       two important private initiatives with huge
McCain’s experience highlights the importance          potential for transformative impact are the
of the policy space and gestation period for           “e-Choupal” of ITC and the mobile SMS based
establishing a viable business. It undertook           information service by Reuters Market Light
R&D experiments in five states to grow the             (RML). The “e-choupal” started in 2000 and
“right” quality of potato for its core business        seeks to connect rural India with data access to
and was successful in establishing a reliable          market prices, agricultural extension services,
supply chain for French fries for the fast-            and other e-services. The RML (Reuters Market
evolving QSR market. Its produce effectively           Light) information service, stated in 2007,
substituted for imports for the highly specific        provides a highly customized and localized
types of French fries demanded by the QSR              agricultural and related information service,
giant, McDonald’s. Similarly, PepsiCo developed        primarily aimed at farmers, through mobile
the specific variety of potato needed for its Frito    phone-based text messages.
Lays chips business, and it now operates a large
“contract” farming program for an assured              The interest of the private sector in the 1990s
supply for its factories. The demand from the          was triggered by the economic reforms and
burgeoning snack/fast-food market is one key           expectations of rapid increases in consumer
factor behind the “success” of the potato supply       income. There was also an expectation that the
chains (including the wide network of much-            well-known challenges in making agriculture
needed cold stores for potatoes) despite existing      more competitive (including critical lower-level
obstacles.                                             policy reforms to address the complex and
                                                       myriad state and central laws on agriculture,
The change in the policy for wholesale trade           taxation, marketing, access to land, roads,
induced investments such as METRO Cash &               power, and water) would be progressively
Carry in Karnataka (1997). The DCM Shriram             addressed, even if only slowly. Among these,
(DSCL) group launched its Haryali Kissan Bazaar        market regulations (such as the APMC Act, ECA,
(rural markets) in north India in 2002, and            and lack of a uniform national tax code for
FieldFresh Foods started to market horticultural       processed foods) are particularly constraining to
products in 2004. The new seed policy attracted        the development of efficient agricultural value
a number of multinational companies in the             chains, raising transaction costs and creating
seed sector, like Novartis (later to become            uncertainty over whether private investors can
Syngenta), Cargill, and Pioneer Seeds. Private         sustain their investments and remain viable.



                                                                                                     191
Republic of India: Accelerating Agricultural Productivity Growth




Small and medium businesses are particularly       agri-business growth is demonstrated by
severely affected by these circumstances, as       the development of horticultural exports
they have much less capacity (either in terms of   (especially grapes) and rapid growth of the
profit margins or scale of operations) to absorb   poultry and dairy processing industries. On the
the costs as the large corporate entities may be   export front, the government’s Agricultural
able to do.                                        and Processed Food Products Export
                                                   Development Authority (APEDA), established in
In reality, several large and visible corporate    1985 with a mandate to develop and promote
initiatives were rooted in a mix of strategic      exports of agricultural and processed food
interests, public relations, or corporate social   products from India, has been successful
responsibility activities. Some initiatives        through sustained efforts in improving
have matured into well-integrated and              the transparency of supply, implementing
managed supply chains for the investors’           traceability and quality standards, and
main businesses. Others have remained small,       negotiating entry into the European Union
reflecting the corporate social responsibility     market for Indian grapes, and Japan and the
orientation, or have failed to achieve             United States for fresh mangoes.
commercial viability. For example, many
Indian corporations have attempted chain           Other successful experiences also point to
store grocery retailing but have not seen          the important role public policy can play in
any significant success. The development           market development. For example, rapidly
of the cold chain infrastructure is a case in      rising domestic incomes and urbanization, and
point. Cold chains are a priority for moving       the associated growth in demand for animal
perishables from their production base to          proteins and milk products, have helped
large consumption centers in urban India.          the poultry industry, which has integrated
Despite government incentives to develop the       vertically to benefit from economies of scale
sector, however, the poor infrastructure (power    and value chain integration. The industry has
and roads) and the regulatory framework for        benefited significantly by the absence of any
agricultural markets have hampered investment      truly inimical regulations to hinder business
and growth in cold chains. Where cold stores       development, but its growth has been limited
have emerged, they have been driven by             by the lack of proper cold chain infrastructure
strong demand (and high margins) for specific      for frozen and processed poultry meat.
commodities such as potatoes, which have           Similarly, the removal of market restrictions
made the investment viable despite existing        in the dairy industry led to a rapid response by
hindrances and transaction costs. They have not    the private sector in creating new value chains
emerged as a supply-led solution.                  in many parts of the country. The projected
                                                   growth in demand is now attracting large
The importance of public support and               international and domestic investments in
appropriate policies to promote rapid              the sector.




 192
                                       Annex 2: land use, grain
annexes                                stocks and the structure of
                                       production

 Figure A.2.1: Food grain area, production, and yields and irrigated area
            250                                                                                                                                                                                                                                                                                                                50

                                                                                                                                                                                                                                                                                                                               45

            200                                                                                                                                                                                                                                                                                                                40

                                                                                                                                                                                                                                                                                                                               35

            150                                                                                                                                                                                                                                                                                                                30

                                                                                                                                                                                                                                                                                                                               25

            100                                                                                                                                                                                                                                                                                                                20

                                                                                                                                                                                                                                                                                                                               15

              50                                                                                                                                                                                                                                                                                                               10

                                                                                                                                                                                                                                                                                                                               5

               0                                                                                                                                                                                                                                                                                                               0
                   1950-51
                             1952-53
                                       1954-55
                                                 1956-57
                                                           1958-59
                                                                     1960-61
                                                                               1962-63
                                                                                         1964-65
                                                                                                   1966-67
                                                                                                             1968-69
                                                                                                                       1970-71
                                                                                                                                 1972-73
                                                                                                                                           1974-75
                                                                                                                                                     1976-77
                                                                                                                                                               1978-79
                                                                                                                                                                         1980-81
                                                                                                                                                                                   1982-83
                                                                                                                                                                                             1984-85
                                                                                                                                                                                                       1986-87
                                                                                                                                                                                                                 1988-89
                                                                                                                                                                                                                           1990-91
                                                                                                                                                                                                                                     1992-93
                                                                                                                                                                                                                                               1994-95
                                                                                                                                                                                                                                                         1996-97
                                                                                                                                                                                                                                                                   1998-99
                                                                                                                                                                                                                                                                             2000-01
                                                                                                                                                                                                                                                                                       2002-03
                                                                                                                                                                                                                                                                                                 2004-05
                                                                                                                                                                                                                                                                                                           2006-07
                                                                                                                                                                                                                                                                                                                     2008-09
                                                                                            Area                                             Production                                                             Area Irrigated (%)
 Source: 
         Authors, using CSO data.

 Figure A.2.2: Trends in agricultural land use
            200

            180

            160

            140

            120

            100

             80

             60

             40

             20

              0
                              1950-51       1960-61       1970-71                                                                                                 1980-81                                          1990-91                                     2000-01        2008-09
                             Tree Crops/Groves     Perm. Pasture                                                                                               Fallow Land                                         NSA                           GSA                Net Irrigated Area
 Source: 
         Authors, using CSO data.


                                                                                                                                                                                                                                                                                                                                    193
Republic of India: Accelerating Agricultural Productivity Growth




Figure A.2.3: Stocks and procurement as percentage of production (right)

                    45.0%                                                                                                                                                                         80.00

                    40.0%                                                                                                                                                                         70.00

                    35.0%
                                                                                                                                                                                                  60.00

                    30.0%
                                                                                                                                                                                                  50.00
                    25.0%
                                                                                                                                                                                                  40.00
                    20.0%
                                                                                                                                                                                                  30.00
                    15.0%

                                                                                                                                                                                                  20.00
                    10.0%

                     5.0%                                                                                                                                                                         10.00


                     0.0%                                                                                                                                                                         0.00
                              1997-98

                                        1998-99

                                                   1999-00

                                                              2000-01

                                                                         2001-02

                                                                                   2002-03

                                                                                             2003-04

                                                                                                       2004-05

                                                                                                                 2005-06

                                                                                                                           2006-07

                                                                                                                                     2007-08

                                                                                                                                               2008-09

                                                                                                                                                          2009-10

                                                                                                                                                                    2010-11

                                                                                                                                                                              2011-12

                                                                                                                                                                                        2012-13
                       Stocks at Jan 1                                  Wheat Procurement (%)                                                            Rice Procurement (%)


Source: 
        Authors, using FCI and CSO data.


Figure A.2.4: Changes in land allocation, 1982/83–2007/08


                    100

                     90                           1.8
                                        1.3                  2.3                                        2.5
                                                                                                                 3.8
                                                  5.2                                                                1.6
                     80                                                                                          5.2                                                   Sugarcane
                                                  9.4
                     70                                                                                                                                                Fruits & Vegs
                                                                                                                 14.9
                                              14.3                                                                                                                     Beverages & Spices
                     60
                                                                                                                 11.9                                                  Fibres
                     50
                                              21.9                                                               14.3                                                  Oilseeds
                     40
                                                                                                                                                                       Pulses
                     30                       13.1                                                               14.3                                                  Other cereals

                     20                                                                                                                                                Wheat
                                              22.6                                                               22.1                                                  Rice
                     10

                      0
                                    TE1982/83                                                          TE2007/08

Source: 
        Authors, using CSO data.




  194
                            Annex 3: Comparing the
annexes                     Impact of Alternative Rainfall
                            Estimates


  Figure A.3.1: Comparison of rainfall anomalies using IMD and district (ICRISAT/NCAP) data

                      20


                      10


                       0


                       -
                      10


                       -
                      20


                       -
                      30

                            1970               1980                    1990              2000                  2010
                                                                  Year
                                           Aggregate Rainfall Anomaly (%) : Deviation from District Avg.
                                           Aggregate Rainfall Anomaly (%) : IMD
  Source: 
          Kshirsagar and Gautam 2013.
  Note: 
        District-level rainfall data are for annual rainfall and IMD rainfall data are for aggregate national-level monsoon
        rainfall (i.e., for June–September of each year). The observed differences thus are due to differences in the rainfall
        period and geographical coverage, as the district-level analysis is based on the original (as they existed in 1970) 207
        districts with available rainfall data.

  Figure A.3.2: Relationship between agricultural productivity and rainfall (using district rainfall data)

               20


               10


                0


              - 10


              - 20


                     1970                    1980                      1990                  2000                     2010
                                                                       Year
                                   Avg. Rainfall Anomaly (%) : IMD             Trend-Deviation in Real Ag. Production
                                   Trend - Deviation in Cropped Area           Trend-Deviation in Real Ag. Productivity
  Source: 
          Kshirsagar and Gautam 2013.


                                                                                                                             195
Republic of India: Accelerating Agricultural Productivity Growth




Figure A.3.3: Cereal and noncereal productivity deviations and rainfall anomalies


                           20                                                                                                                  20




                                                                                                                                                           Trend Deviation in Productivity
                           10
   Rainfall Anomaly (%)




                                                                                                                                               10


                            0                                                                                                                  0


                           -10                                                                                                                 -10


                           -20
                                                                                                                                               -20
                                 1970                 1980                       1990                        2000                       2010
                                                                                 Year
                                        Avg. Rainfall Anomaly (%)                         Trend-Deviation in Real Cereal Productivity
                                        Trend-Deviation in Real Non-Cereal Productivity


Source: 
        Kshirsagar and Gautam 2013.




Figure A.3.4: Sensitivity of noncereal area share to rainfall




                                                                                                                                                        Trend-Deviation in Non-Cereal Area Share
                           20                                                                                                                      4



                           10
   Avg. Rainfall Anomaly




                                                                                                                                                   2


                            0                                                                                                                      0


                           -10
                                                                                                                                                   -2


                           -20
                                                                                                                                                   -4
                                 1970                  1980                      1990                         2000                      2010
                                                                                 Year
                                                Avg. Rainfall Anomaly (%)        Trend-Deviation in Non-Cereal Area Share


Source: 
        Kshirsagar and Gautam 2013.




  196
                      Annex 4: Sub-National
annexes               Heterogeneity in Agricultural
                      Productivity



 Table A.4.1:  Disparities in productivity within states

                                      Coefficient of variation                    Gini coefficient
  State                     TE 1991         TE 2001         TE 2007     TE 1991       TE 2001        TE 2007
  Andhra Pradesh              45.97           46.79          40.95       0.25           0.24          0.22
  Assam*                      21.46           19.66          22.43       0.12           0.11          0.12
  Bihar*                      19.16           26.23          27.24       0.11           0.14          0.14
  Chhattisgarh*               19.00           15.24          27.41       0.09           0.08          0.14
  Gujarat                     35.98           32.92          29.69       0.19           0.17          0.16
  Haryana                     45.94           38.95          42.19       0.19           0.17          0.16
  Himachal Pradesh            29.06           34.69          35.21       0.15           0.19          0.18
  Jharkhand*                  45.37           34.3           60.33       0.23           0.18          0.27
  Karnataka                   60.3            48.88          45.74       0.27           0.23          0.23
  Kerala                      36.61           29.79          31.87        0.2           0.16          0.17
  Madhya Pradesh*             25.18           27.05          28.18       0.14           0.15          0.15
  Maharashtra                 52.02           52.83          56.56       0.25           0.26          0.26
  Odisha*                     25.67           26.37          18.26       0.14           0.14           0.1
  Punjab                      19.98           18.76          19.49       0.11           0.1           0.11
  Rajasthan*                  53.09           60.77          53.27        0.3           0.34           0.3
  Tamil Nadu                  47.23           36.58          32.72       0.24           0.2           0.18
  Uttar Pradesh*              23.01           23.56          27.47       0.13           0.13          0.15
  Uttarakhand*                30.47           52.28              52.7    0.12           0.23          0.24
  West Bengal                 37.85           37.08          32.72       0.21           0.2           0.17
  Overall                     58.31           59.79          56.85       0.30           0.33          0.29
 Source:  Kumar and Jain 2012.
 Note:  * indicates Low-income States.




                                                                                                             197
Republic of India: Accelerating Agricultural Productivity Growth




Table A.4.2: Districts’ transition using irrigation typology
                                                                                                  2000s
1970s                                                         <25%                     25-50%                      >50%                       Total
<25%                                                             89                        64                         19                       172
                                                               (30)                      (22)                        (6)                       (58)
25-50%                                                            3                        26                         49                         78
                                                                (1)                        (9)                      (17)                       (26)
>50%                                                              1                          1                        43                         45
                                                                (0)                        (0)                      (15)                       (15)
Total                                                            93                        91                       111                        295
                                                               (32)                      (31)                       (38)                     (100)
Source:  Authors, using ICRISAT-NCAP district database.
Note:  
       Irrigation typology defined as indicated percentage of net irrigated area in a district.
       Parentheses indicate the percentage of all districts in each category.

               verage level of productivity, rainfall, and population density for districts categorized by
Figure A.4.1: A
              level of irrigation intensity for TE 2007
                                        2000                                                                                                  50




                                                                                                                                                      Thousand Rupees/Ha (Net cropped area)
                                        1800                                                                                                  45
        Rainfall in mm and population




                                        1600                                                                                                  40
                                        1400                                                                                                  35
                 density (x200)




                                        1200                                                                                                  30
                                        1000                                                                                                  25
                                         800                                                                                                  20
                                         600                                                                                                  15

                                         400                                                                                                  10

                                         200                                                                                                  5

                                            0                                                                                                 0
                                                      0                0-9%             9-39%             39-68%               68%+
                                                   Land Productivity    Labor Productivity       Population Density        Annual Rainfall
Source: Authors, using ICRISAT-NCAP district database.
Note:  
       Less-irrigated districts are often endowed with higher rainfall and hence may be as or more productive than
       districts with significantly higher levels of irrigation development.

               eal value of crop production across district typologies
Figure A.4.2: R
                                        50,000
           Crop Production per ha




                                        40,000
               (2005-2007 Rs.)




                                        30,000
                                        20,000
                                        10,000
                                            0
                                                 70s 80s 90s 00s          70s 80s 90s 00s         70s 80s 90s 00s              70s 80s 90s 00s
                                                 High Yield - Growth       High Yield - Stable     Low Yield - Growth           Low Yield - Stable
Source: Kshirsagar and Gautam 2013.
Note:  
       Total value of crop production includes crops other than cereals but not fruits, vegetables, and spices, for which
       production data are not available.


  198
                                                                     ﻿Annex 4: Sub-National Heterogeneity in Agricultural Productivity




Table A.4.3: District growth typologies by agro-ecological zone

                                                    High Yield/        High Yield/        Low Yield /         Low Yield/
Agro-ecological zone                                 Growth              Stable            Growth               Stable                Total
Humid                                                      3                 32                17                  39                    91
Semi-Arid Temperate                                        5                 12                42                  12                    71
Semi-Arid Tropical                                         4                 13                34                  67                   118
Arid                                                       1                  2                10                    4                   17
Total                                                    13                 59                103                 122                   297
Source: Kshirsagar and Gautam 2013.

               verage rainfall across district typologies by decade
Figure A.4.3: A
                           1,500
    Annual Rainfall (mm)




                           1,000


                             500


                                 0
                                       70s 80s 90s 00s            70s 80s 90s 00s          70s 80s 90s 00s            70s 80s 90s 00s
                                        High Yield - Growth         High Yield - Stable      Low Yield - Growth         Low Yield - Stable

Source: Kshirsagar and Gautam 2013.


               verage rainfall anomaly across district typologies by decade
Figure A.4.4: A
    Annual Rainfall (mm)




                            5

                            0

                            -5

                           -10
                                     70s  80s     90s   00s   70s  80s     90s    00s     70s   80s     90s  00s     70s   80s    90s     00s
                                      High Yield - Growth       High Yield - Stable         Low Yield - Growth          Low Yield - Stable

Source: Kshirsagar and Gautam 2013.


Table A.4.4:  District growth typologies by irrigation development
                                                                      Dynamic Irrigation Typology
LR growth typology                                                    Low         Moderate/ growth                 High                  Total
High yield–growth                                                        1                  0                        12                    13
High yield–stable                                                       20                  6                        33                    59
Low yield–growth                                                        18                 35                        50                   103
Low yield–stable                                                        50                 42                        28                   120
Total                                                                   89                 83                       123                   295
Source:  Kshirsagar and Gautam 2013.
Note:  
       Category “low” refers to districts with less than 25% irrigated area in both the 1970s and 2000s; “high” refers to
       districts with more than 50% irrigated area in both the 1970s and 2000s; “moderate/growth” refers to districts that
       had less than 50% irrigated area in the 1970s but subsequently experienced growth in irrigated area.



                                                                                                                                                199
Republic of India: Accelerating Agricultural Productivity Growth




               hanging patterns in modes of irrigation across typologies
Figure A.4.5: C

                 60
        Pumps




                 40


                 20


                  0
                      70s 80s 90s 00s       70s 80s 90s 00s        70s 80s 90s 00s      70s 80s 90s 00s
                      High Yield - Growth    High Yield - Stable   Low Yield - Growth    Low Yield - Stable


            150


            100
        Wells




                 50


                  0
                      70s 80s 90s 00s       70s 80s 90s 00s        70s 80s 90s 00s      70s 80s 90s 00s
                      High Yield - Growth    High Yield - Stable   Low Yield - Growth    Low Yield - Stable


                 6


                 4
        Tanks




                 2


                 0

                      70s 80s 90s 00s       70s 80s 90s 00s        70s 80s 90s 00s      70s 80s 90s 00s
                      High Yield - Growth    High Yield - Stable   Low Yield - Growth    Low Yield - Stable


            150


            100
        Canals




                 50


                 0

                      70s 80s 90s 00s       70s 80s 90s 00s        70s 80s 90s 00s      70s 80s 90s 00s
                      High Yield - Growth    High Yield - Stable   Low Yield - Growth    Low Yield - Stable

Source: Kshirsagar and Gautam 2013.
Note:  
       The top panel shows the median number of pumps (diesel and electric) across typologies and decades. The bottom
       three panels show the median area irrigated using wells, tanks, and canals across typologies and decades.




  200
                                                                               ﻿Annex 4: Sub-National Heterogeneity in Agricultural Productivity




               istrict productivity growth and convergence
Figure A.4.6: D

                                                     Growth in cereal yields                                                                  Growth in crop productivity
                                   2                                                                                         3




                                                                                               Growth in Crop Productivity
     Cereal Productivity Growth




                                                                                                                             2
                                  1.5

                                                                                                                             1
                                   1
                                                                                                                             0
                                   .5
                                                                                                                             -1

                                   0
                                                                                                                             -2
                                        0                1                 2               3                                      0        10000       20000          30000    40000
                                                             1970 Yields                                                                         Productivity Level in 1970
                                        bandwidth = .8                                                                            bandwidth = .8

Source: Kshirsagar and Gautam 2013.


Table A.4.5:  Breakdown of agro-ecological zones by state
                                                                                  Semi-Arid                                           Semi-Arid
State                                                        Humid                Temperate                                            Tropical            Arid               Total
Andhra Pradesh                                                   5                     0                                                 14                   1               20
Assam                                                           10                     0                                                  0                   0               10
Bihar                                                            5                     2                                                  4                   0               11
Gujarat                                                          0                     0                                                 17                   1               18
Haryana                                                          0                     5                                                  0                   2                 7
Himachal Pradesh                                                 9                     0                                                  0                   0                 9
Karnataka                                                        6                     0                                                 12                   1               19
Kerala                                                          10                     0                                                  0                   0               10
Madhya Pradesh                                                   5                     9                                                 29                   0               43
Maharashtra                                                      6                     0                                                 19                   0               25
Odisha                                                          11                     0                                                  2                   0               13
Punjab                                                           0                     8                                                  0                   3               11
Rajasthan                                                        0                     5                                                 12                   9                26
Tamil Nadu                                                       3                     0                                                  9                   0                12
Uttar Pradesh                                                    7                    41                                                  0                   0                48
West Bengal                                                     14                     1                                                  0                   0                15
 Total                                                          91                    71                                                118                 17                297
Source:  Kshirsagar and Gautam 2013.
Note:  Own calculations using broad agro-zone classification of districts by ICRISAT.




                                                                                                                                                                                      201
Republic of India: Accelerating Agricultural Productivity Growth




Table A.4.6:  Levels and changes in the composition of crop area allocations
 Area
 share       High Yield/Growth                     High Yield/Stable    Low Yield/Growth                         Low Yield/Stable
 Decade                     %                                       %                  %                                         %
 averages 1970s 2000s change                    1970s 2000s change 1970s 2000s change                         1970s 2000s change
 Cereals    0.60    0.55    -8                   0.66    0.60      -9 0.61   0.53     -14                      0.63   0.56      -11
 Pulses     0.11    0.08 -31                     0.07    0.05    -25  0.17   0.11     -33                      0.15   0.16        8
 Oils       0.04    0.03 -18                     0.09    0.15     59  0.10   0.19      94                      0.09   0.14       53
 Sugar      0.07    0.09   36                    0.02    0.03     50  0.02   0.03      21                      0.01   0.02       98
 Cotton     0.03    0.06   86                    0.02    0.01    -34  0.03   0.05      37                      0.06   0.04      -25
 Fruits/veg 0.06    0.10   70                    0.09    0.14     44  0.02   0.04    173                       0.03   0.06     105
 Onions     0.00    0.00     7                   0.00    0.00     15  0.00   0.00    120                       0.00   0.00     103
 Potatoes   0.03    0.02 -32                     0.01    0.01     56  0.01   0.01    102                       0.00   0.01     101
 Fodder     0.06    0.07     4                   0.05    0.03    -40  0.05   0.05       -4                     0.04   0.03      -11
 Total       1.0     1.0     0                    1.0     1.0       0  1.0    1.0        0                      1.0    1.0        0
Source:  Kshirsagar and Gautam 2013.


Table A.4.7: Correlates of real cereal productivity
                     Dependent Variable : Real Valu of Cereal Production per ha (2005-2007 Rs)
                                             High Yield -       High Yield -       Low Yield -                              Low Yield -
 Sample                         Full            Growth            Stable            Growth                                    Stable
 Lag Dep. Variable            0.221***          0.360***            0.418***          0.145***                                0.126***
                               0.0449            0.0576              0.0461            0.0316                                  0.0403
 Annual Rainfall              0.185***           0.0322              0.0418           0.212***                                0.263***
                               0.0314            0.0506              0.0252            0.0596                                  0.0347
 Share Irrigated             0.0745***             0.209             0.0395           0.0731**                                 0.0348
                               0.0227              0.108             0.0322            0.0233                                    0.025
 Share HYV                   0.0214***       0.00742***            0.0235**           0.00994                                0.0251***
                              0.00448           0.00122            0.00763            0.00792                                 0.00687
 Fertilizer Intensity        0.0653***          0.0929*             0.0341*          0.0620***                                0.0455*
                               0.0126            0.0435              0.0156            0.0162                                  0.0218
 Road Density                 0.00394            0.0126            0.0523**            0.0265                                 0.00997
                                 0.047           0.0626              0.0203            0.0618                                  0.0399
 Constant                     5.984***          6.452***            5.267***          6.706***                                6.288***
                                 0.404              0.25              0.434              0.311                                   0.486
 Observations                    7,347               328              1,561              2,435                                   3,023
 R-squared                       0.576             0.852              0.686              0,696                                   0.481
 Number of Districts                      286                   13                       58                    103                   112
Source: Kshirsagar and Gautam 2013.
Note: 
      All variables are in logs (i.e., all the coefficients are elasticities). All regressions use both district fixed effects and year
      dummies. Robust standard errors clustered at the state level: *** p<0.01, ** p<0.05, * p<0.1.




  202
                                                     ﻿Annex 4: Sub-National Heterogeneity in Agricultural Productivity




Table A.4.8: Correlates of real agricultural productivity
                 Dependent Variable : Real Value of Agricultural Production per ha (2005–2007 Rs)
                                             High Yield -      High Yield -       Low Yield -     Low Yield -
 Sample                         Full           Growth             Stable            Growth          Stable
 Lag Dep. Variable            0.368***         00351**              0.806***           0.203**      0.254***
                                 0.098          0.0952               0.0985            0.0689          0.043
 Annual Rainfall              0.191***          0.0198               0.0333           0.240***      0.263***
                                 0.033          0.0429               0.0295            0.0618          0.038
 Share Irrigated              0.0856*            0.290*              0.0406            0.126**       0.0633
                               0.0399             0.131              0.0317            0.0376        0.0393
 Share HYV                    0.00315          0.00601              0.00968            0.0143       0.00281
                              0.00938          0.00906              0.00621            0.0132       0.00874
 Fertilizer Intensity        0.0954***         0.0816**            0.0831***         0.0870**      0.0711***
                               0.0219           0.0209               0.0176            0.0306        0.0195
 Road Density                  0.0449            0.198*              0.0262            0.0415            0.01
                               0.0392           0.0781                 0.018             0.101       0.0327
 Constant                     4.341***         5.860***                 1.34          5.909***      4.942***
                                 0.967            0.891                0.876             0.364          0421
 Observations                    7,333              328                1,561            2,435          3,009
 R-squared                       0.571            0.816                0.734              0.65          0563
 Number of Districts               286               13                   58               103           112
Source: Kshirsagar and Gautam 2013.
Note: 
      All variables are in logs (i.e., all the coefficients are elasticities). All regressions use both district fixed effects and year
      dummies. Robust standard errors clustered at the state level: *** p<0.01, ** p<0.05, * p<0.1.

Table A.4.9: Correlates of real non-Cereal productivity
                 Dependent Variable : Real Value of Non-Cereal Production per HA (2005-2007 Rs)
                                             High Yield -     High Yield -      Low Yield -                                 Low Yield -
 Sample                        Full            Growth            Stable           Growth                                      Stable
 Lag Dep. Variable           0.604***          0.522***            0.796***         0.389***                                  0.471***
                               0.0942             0.124             0.0488            0.112                                    0.0578
 Annual Rainfall             0.144***            0.117*             0.0402          0.207***                                   0.144**
                               0.0284             0.048             0.0238           0.0545                                    0.0509
 Share Irrigated               0.0909           0.576**             0.0444           0.183**                                  0.0831*
                                0.057             0.163               0.125          0.0791                                    0.0428
 Share HYV                     0.0197           0.0287              0.0536         0.00498                                    0.0339*
                               0.0124           0.0155              0.0364           0.0207                                    0.0174
 Fertilizer Intensity       0.0776***           0.0494              0.0913         0.0880**                                    0.0318
                               0.0246           0.0417                0.058          0.0275                                    0.0347
 Road Density                  0.0141            0.369*             0.103**          0.0268                                   0.00494
                               0.0307              0.18             0.0338           0.0988                                    0.0253
 Constant                     2.296**           3.720**              1.175*         4.114***                                  3.487***
                                0.943            -1.371               0.632            0.862                                     0.492
 Observations                   7.294               316               1.531            2.429                                     3.018
 R-squared                      0.488             0.616               0.687            0.459                                     0.416
 Number of Districts              286                13                  58              103                                       112
Source: Kshirsagar and Gautam 2013.
Note: 
      All variables are in logs (i.e., all the coefficients are elasticities). All regressions use both district fixed effects and year
      dummies. Robust standard errors clustered at the state level: *** p<0.01, ** p<0.05, * p<0.1.


                                                                                                                                          203
Republic of India: Accelerating Agricultural Productivity Growth




Table A.4.10: Cross-sectional correlates of real cereal productivity
                   Dependent Variable : Real Value of Cereal Production per ha (2005-2007 Rs)
Sample                               Full                Full                Full                                  Full
Mean Rainfall                      0.315***            0.343***              0.325***                              0.309***
                                    0.0698              0.0669                0.0875                                0.0763
Norm. Std. Dev of Rainfall        -0.145***           -0.147***             -0.148***                             -0.142***
                                     0.046              0.0399                0.0443                                0.0376
Share Irrigated                    0.220***             0.0375               0.175***                               0.0313
                                    0.0377              0.0354                0.0373                                0.0324
Share HYN                         0.0963**             0.245***              0.142***                              0.247***
                                    0.0373              0.0665                0.0393                                0.0626
Fertilizer Intensity               0.241***            0.209***              0.244***                              0.224***
                                    0.0399              0.0391                0.0389                                0.0405
Markets pc                        0.0714**              0.0308               0.0695**                               0.0335
                                    0.0349              0.0332                0.0343                                0.0335
Road Density                        0.0156              0.0641               0.00746                                0.0591
                                    0.0308              0.0479                  0.035                               0.0503
Constant                           7.911***            7.780***              7.874***                              8.008***
                                     0.645               0.606                  0.708                                0.626
State Dummies                           No                 Yes                     No                                  Yes
AEZ Dummies                                  No                    No                         Yes                      Yes
Observations                                275                   275                         275                      275
R-squared                                 0.723                 0.848                       0.753                    0.855
Source: Kshirsagar and Gautam 2013.
Note: 
      All variables are in logs (i.e., all the coefficients are elasticities). All regressions estimate the (cross-sectional)
      relationship between mean values across districts. Normalized std. deviation of rainfall measures the standard
      deviation of rainfall anomalies across the sample period (i.e., 1970–2007). Robust standard errors: *** p<0.01,
      ** p<0.05, * p<0.1.




  204
annexes
                     Annex 5: Changing Structure
                     of Household Incomes in Bihar



 Table A.5.1.  Distribution of income by source and class, 2011 (%)
                      Own                                  Non-
                   production                           agricultural    Non-
 Caste/                 in     Agricultural     All         own      agricultural Regular Other
 community         agriculture   wages      agriculture production      wages employment income Remittances
 Agricultural        16.4           8.9       25.3          5.4        15.3         3.9    16.1   34.0
 labor
 Poor peasant        24.3           1.5        25.8         25.4      4.2       5.6      11.2     27.8
 Middle peasant      38.4           1.8        40.2         19.5      7.0       1.6       9.8     22.0
 Big peasant         34.8           0.1        34.9          9.3      1.5      17.5      18.3     18.5
 Landlord/           11.1           0.0        11.1          0.7      0.0      44.7      20.5     23.0
 supervision
 Nonagricultural     10.1           1.2        11.3          4.2      5.2      27.9      16.3     35.0
 wages
 Nonagricultural      9.5           0.6        10.1         49.9      1.5       0.2      19.6     18.7
 self-employment
 All total           22.3           4.2        26.5         10.7      8.2      10.6      16.0     28.0
 Source:  IHD Bihar Survey 2011; Rodgers and Sharma 2011.




                                                                                                      205
                        Annex 6: Growth
annexes                 Decomposition: Methodology
                        and Selected Indicators



 Growth decomposition: Data and                                  Data on the value of the output of the selected
 methodology                                                     crops (for generating implicit prices of the
                                                                 selected commodities) were obtained from
 Data. Data on major crops grown in 20                           the National Accounts Statistics published
 major agricultural states1891 and data on area,                 by CSO (Ministry of Statistics and Program
 production, and yield of important crops                        Implementation, Government of India). The
 were compiled from the Indian Agricultural                      price of a commodity was obtained by dividing
 Statistics, Agricultural Statistics at Glance, and              its value of output (at current prices) by its level
 Horticultural Database, all published by the                    of production. The current prices were deflated
 Ministry of Agriculture, Government of India:                   by the general WPI to convert these into real
 cereals (rice, wheat, maize, sorghum, pearl                     prices (at 1993–94 base). The data series on
 millet, finger millet, barley, small millets);                  area, production, and prices were smoothened
 pulses (chickpea, pigeonpeas, and other pulses);                by applying a Hodrick-Prescott (HP) filtering1902
 oilseeds (groundnut, sesamum, rapeseed-                         technique with an adjustment factor of 6.25.
 mustard, soybean, linseed, sunflower, safflower,                The HP filtered data series were used for
 castor, and niger seed); fibers (cotton, jute, and              analyzing the patterns and sources of growth.
 sunhemp); spices (arecanut, cardamom, chillies,
 pepper, turmeric, ginger, garlic, and coriander);               Methodology. Following Minot et al. (2006),
 fruits (bananas, cashewnuts, and other fruits);                 the change in gross revenue from each crop can
 vegetables (potatoes, sweet potatoes, onions,                   be decomposed into (i) change in cropped area,
 tapioca, and other vegetables); beverages                       (ii) change in yield, (iii) change in real price,
 (tea and coffee); coconut, sugarcane, tobacco,                  and (iv) a residual representing interaction
 rubber, and cluster bean. These data were                       among (i) to (iii). When more than one crop
 supplemented with the data from state-specific                  is considered, an additional source of change
 Statistical Abstracts published by the state                    is the reallocation of area between crops, or
 governments. The crops included account for                     (v) diversification. Typically, assuming profit
 more than 90 percent of the cropped area and                    maximization behaviors, farmers are expected
 the value of the output of the crop sector.
                                                                 190	The HP filter is a data-smoothening technique commonly
                                                                     applied to remove short-term fluctuations from time-series
 189	 Andhra Pradesh, Assam, Bihar, Chhattisgarh, Gujarat,           data. It generates a smoothened nonlinear representation of
      Haryana, Himachal Pradesh, Jharkhand, Karnataka, Kerala,       a time series. The adjustment of the sensitivity of the trend
      Madhya Pradesh, Maharashtra, Odisha, Punjab, Rajasthan,        to short-term fluctuations is done by applying a suitable
      Tamil Nadu, Uttarakhand, Uttar Pradesh, and West Bengal.       adjustment factor.




   206
                                   ﻿Annex 6: Growth Decomposition: Methodology and Selected Indicators




to diversify from lower- to higher-value crops.
But based on their individual circumstances and
in any particular year, farmers may choose to
diversify to less risky crops that may be of lower
value.                                                                                                    (5)

Assuming that farmers behave rationally, the                                           
individual farmer maximizes profits from land
by choosing a production mix, inputs, and                 Equation (5) decomposes change in gross
technologies subject to the individual’s resource         revenue due to change (i) in total cropped area,
endowments and markets. If Ai is area under               (ii) crop yields or technology, (iii) real prices,
crop i, Yi is its production per unit area, and Pi is     and (iv) land reallocation or diversification.
the real price per unit of production, then the           The first term on the right-hand side of this
gross revenue R from n crops can be written as:           equation represents the change in gross
                                                          revenue due to change in total cropped area.
              	                                    (1)   The expression           is the weighted average

Expressing A1 as the share of crop i in the total         of the gross revenue per hectare, the weights
                                                          being share of each crop (ai) in the total cropped
cropped area,               , and substituting in         area. The second term on the right-hand side
equation (1):                                             captures change in the gross revenue due to a
                                                    (2)   change in the real prices of commodities. The
                                                         third term measures change in gross revenue
                                                          due to changes in crop yields or technology.
Total derivatives of both sides of equation (2)           The fourth term represents the change in gross
provides absolute contribution of changes                 revenue associated with changes in the crop
in these components to the change in gross                composition. If the fourth term is positive, this
revenue:                                                  indicates a reallocation of land from lower-
                                                          value to higher-value crops. Dividing both
                                                   (3)   sides of equation (5) by the overall change in
                                                          gross revenue (dR) gives the proportionate
                                                          contribution of each source to the overall
Equation (3) is only an approximation, as it              change in gross revenue or agricultural growth.
excludes the interaction term. The second term
on the right-hand side of this equation can be            Description of the study regions
further decomposed from a change in sums to
the sum of changes, as follows:                           Northern region, excluding the hill states
                                                          of Himachal Pradesh, and Uttarakhand, has
                                                          alluvial soils and a semi-arid to humid climate.
                                               (4)
                                                         Agriculture contributes about one-fifth to
And, further expansion of the second term of              GDP in this region. In the plains, irrigation
equation (4) results in the following expression:         infrastructure is well developed, with more



                                                                                                        207
Republic of India: Accelerating Agricultural Productivity Growth




than 80 percent of cropped area under               capita income. Economic dependence on
irrigation. The average size of land holding is     agriculture is the lowest in this region. The
small. The land is intensively cultivated, with     region receives less rainfall than the others,
wheat and rice as the main crops in the plains,     and agriculture is highly diversified, with
and maize, wheat, fruits, and vegetables in the     oil seeds (mainly groundnuts and rapeseed-
hills. Fertilizer consumption per unit of cropped   mustard), wheat, pulses, fruits, and vegetables
area is one of the highest in this region. Road     as important crops. Coarse cereals, such as
density and telecommunication density are low.      millets and maize, are also widely grown. The
                                                    average land holding is larger than in other
Eastern region is characterized by relatively       regions, but irrigation coverage and fertilizer
high rainfall and widespread cultivation of         use are less.
rice. Rice occupies 56 percent of cropped area.
In value terms, fruits and vegetables outweigh      Southern region, except Kerala, has a semi-
the contribution of rice. Average land holding      arid climate and almost a uniform rainfall
is smaller than in any other region. and the        pattern throughout. Agriculture is largely rain-
level of irrigation development and technology      dependent, but irrigation coverage is better
adoption is also low. The region is heavily         than in other regions. Rice is the dominant
populated but is the least urbanized; it has the    crop, followed by oil seeds, pulses, fruits,
lowest per capita income of all regions. The        vegetables, and spices. The average landholding
region has good road infrastructure. Agriculture    is about 1 hectare. The region has relatively low
accounts for one-fifth of GDP in the region.        population density and is the most urbanized,
                                                    with higher incomes than other regions. The
Western region has low population density           region also has the best road infrastructure and
but high levels of urbanization and per             telecommunication system.




 208
                                 ﻿Annex 6: Growth Decomposition: Methodology and Selected Indicators




Table A.6.1:  Selected regional development indicators
Indicator                                     Northern   Eastern   Western    Southern    All India
Population, in millions, 2011                     282       301      340         251        1210
Geographical area, 000 km2                        667       497     1289         636        3287
Population density, persons/km2, 2011             423       606      263         395          368
Urban population, in %, 2011                      25.1     20.0      35.2        40.9        31.2
Per capita GDP, in Rs (at 2004–05 prices),      26587    21749     38787       43006       35722
2008–10
Share of agriculture in GDP (%), 2004–05 prices
                                    1981–83       43.2     38.0       33.1       33.6        35.1
                                    1991–93       36.1     34.5       25.6       28.3        29.0
                                    1999–01       31.1     28.2       18.8       22.7        24.0
                                    2008–10       22.0     20.8       13.7       15.3        15.7
Gross cropped area, in 000 ha, 2008–10          42765    31968      82940      34476      194301
Gross irrigated area, in %, 2008–10               78.8     43.5       31.0       40.5        45.2
Fertilizer consumption, in kg/ha, 2008–10         173      108          89       180         136
Road density, in km per 000 km , 2008–09
                                 2
                                                  689     1601        548       1552         966
Average size of land holding in ha, 2005–06       0.89     0.69       2.02       1.02        1.23
Tele-density, telephones/100 persons
                                      Urban       86.0    123.9     112.6       151.9      162.1
                                       Rural      23.0     30.1      40.6        46.8       37.9
                                        Total     38.8     48.8      66.0        89.8       76.6
Source:  Birthal et al. 2014.




                                                                                               209
Republic of India: Accelerating Agricultural Productivity Growth




Table A.6.2:  Contribution of crops to agricultural growth: National level
                        Share in gross       Share in real value of     Annual growth in    Share in overall
                         cropped area               output*           real value of output      growth
Crops               1980s 1990s 2000s        1980s 1990s 2000s        1980s 1990s 2000s 1980s 1990s 2000s
Rice                 24.2    24.1    23.6     22.4   21.7      19       3.3    3.1 –0.2    23.1 20.5     –1.7
Wheat                14.2    14.6    15.1      12    12.7     12.6      2.4    5.5     1.2 10.2 20.7      4.6
Maize                 3.5     3.4     4.1      2.1     1.9     2.1      0.5    3.1       5  0.7    1.9    3.2
Other cereals          20    14.6    11.6      5.8     4.1     3.2     –2.7     0.4    1.3 –3.2   –0.2    0.9
Total cereals        61.9    56.7    54.5     42.4    40.4      37        2     3.6    0.7 30.7     43    6.9
Chickpea              4.5       4     3.9      2.8     2.4     2.4      1.2    2.4     5.2  0.7    1.1    3.6
Pigeonpea             1.9       2       2      1.5     1.4       1      3.5    1.7     2.6  1.4    0.1    0.9
Total pulses           14      13    12.5      7.3     6.2     5.2      2.6       1      3  4.6    0.3    4.8
Groundnut             4.5     4.4     3.5      5.2     4.4       3      3.2     –2       2  4.9 –4.2      1.1
Rapeseed and            2     3.3     3.2      2.4     3.1     2.7        9   –1.5     6.1  6.9 –1.7      4.3
mustard
Soybean                0.7       2.8   4.3    0.5        2     2.4     30     8.7    9.1    4.2    3.7    6.6
Other oilseeds         4.3       6.6   7.4    4.3      4.8     4.1     8.7   –2.8    5.4    5.4   –0.5      1
Total oilseeds        10.8      14.2    14   12.4     14.1    12.1     6.9   –0.7    5.4   21.4   –2.6     13
Cotton                 4.5       4.8   4.9    3.9        5       5     1.4    2.8   10.7      4    1.7   14.5
Other fibers           0.7       0.6   0.6    0.7      0.5     0.4     3.7    1.1    1.4    0.8    0.1    0.2
Total fibers           5.2       5.4   5.5    4.7      5.5     5.4     1.7    2.6    9.9    4.8    1.8   14.7
Plantation             1.3       1.6   1.8    1.7      1.8     1.6     5.6    2.7      5    6.8      0    1.1
crops
Spices and             1.1      1.3    1.3    2.8     3.8     3.9      8.5    6.8    3.8    4.8    5.1    5.4
condiments
Fruits                1.4       1.9    2.7    9.4    10.6    14.2      4.4    6.2    5.5   11.3   20.4   24.6
Vegetables            2.1       2.9    3.8    9.8    11.5    13.5      3.6    6.8    6.7     11   19.1   28.9
Sugarcane             1.9       2.3    2.5    8.1     8.6     8.8      1.2      5      0    3.8   13.1   –1.3
Other crops           1.6       1.5    1.9      1     0.9     0.9      0.9      1    6.8    0.8   –0.1      2
All crops             100       100    100    100     100     100      3.1    3.7    3.3    100    100    100
Source:  Birthal et al. 2014.




  210
annexes
                  Annex 7: TFP: Methodology and
                  Crop-Specific Estimates



 A Bird’s-eye view of TFP literature               common thread in these diverse estimates,
 on India                                          however, is that growth in TFP slowed after
                                                   the 1980s. This decline was observed in many
 Since the 1970s, following the Solow (1975)       regions of the country, with output growth
 residual, several studies have estimated TFP in   becoming increasingly input-based (Chand
 Indian agriculture. Most have used the growth     et al. 2010). Most studies have found agricultural
 accounting method, applied at different levels    research to have contributed significantly
 of aggregation; see Rada (2013), Chand, Kumar     and substantially to past growth in TFP, with
 and Kumar (2010), and Kumar and Mittal            significant inter-state and inter-crop variations
 (2006) for reviews of past studies. Because of    in TFP growth.
 data constraints, the majority have been at
 the aggregated sector-wide or national level.
 Some studies have attempted TFP analysis at       TFP estimation: Methodology and
 the individual crop level to overcome potential   data
 problems and complexities associated with
                                                   At the sector-wide level, three recent analyses
 aggregation across crops and inputs, but these
                                                   use the growth accounting framework with
 have primarily been on rice and wheat, with
                                                   Tornqvist-Theil input and output indices to
 concerns about slowing yield growth rates
                                                   measure TFP for a composite output of crops
 and their implications for food security. Most
                                                   and livestock1911: Fuglie (2012), Bosworth and
 analyses were done for the 1980s and 1990s
                                                   Collins (2003), and Rada (2013). Fuglie (2012)
 and are now outdated; they are reviewed briefly
                                                   uses FAO statistics to compile a carefully
 in Kumar and Mittal (2006). Chand, Kumar
                                                   reconciled and comparable database for
 and Kumar (2010) offer the most detailed state
                                                   estimating TFP across 173 countries from 1961
 and crop estimates. They use the CoC data
                                                   to 2009. The study uses an index of real value
 (based on annual surveys) from the Ministry of
                                                   of crops and livestock production covering the
 Agriculture, allowing for better accounting for
                                                   entire agricultural sector in India. Bosworth
 inputs used by farmers than previous studies.
                                                   and Collins (2003) use National Accounts data
 Their analysis stops in 2004/05.
                                                   to estimate TFP in Indian agricultural using
                                                   official GDP estimates from 1961 to 2004 as part
 Depending on the level of aggregation, scope
                                                   of their analysis of the sources of growth in the
 of coverage (for example, crops with or
                                                   Indian economy. These data and TFP estimates
 without livestock, and types of crops covered),
 methodology used, and time periods considered,    191	The advantages of the Tornqvist-Theil index are discussed in
 estimates of TFP growth vary considerably. A          Diewert (1976, 1978).



                                                                                                             211
Republic of India: Accelerating Agricultural Productivity Growth




were updated to 2008 as part of the background                                                                     more appropriately represents a combination
work for World Bank (2012). These estimates                                                                        of technological progress and a change in
use the same growth accounting framework                                                                           the output mix (which, as noted, has been an
as Fuglie but with a different model, inputs,                                                                      important driver of growth in recent years in
parameters, and source of data. A third study                                                                      Indian agriculture). For this study, the DEA
by Rada (2013) uses state and national Value of                                                                    was applied to the same dataset used by Fuglie
Production data a comprehensive set of crops                                                                       (2012) to provide comparable estimates.
and livestock outputs and inputs to arrive at an
independent estimate of TFP for 1980–2008.                                                                         The four TFP estimates provide very similar
                                                                                                                   trends, as shown in Figure A.7.1.
An alternative method of measuring TFP
growth is the Data Envelopment Analysis                                                                            A second level of aggregation is at the state
(DEA) nonparametric methodology. The DEA                                                                           level. Using the National Accounts data on the
allows the two components of TFP, normally                                                                         total value of production, Rada (2013) estimates
referred to as “technical change” and “technical                                                                   TFP using the Tornqvist Theil indices for inputs
efficiency,” to be separated. This procedure                                                                       and outputs for the years from 1980 to 2008.
provides insight into whether the observed                                                                         This analysis uses data on all crops, including
changes in productivity are a result of a shift                                                                    estimated value of output for horticultural crops
in the production frontier or a movement                                                                           and livestock, and cost shares for major inputs
closer to the frontier. A shift in the frontier is                                                                 using aggregated data. A background paper
normally ascribed to technological progress                                                                        for this study also uses the growth accounting
when using physical outputs in the analysis;                                                                       methodology to estimate TFP at the state level
however, when a composite index is used to                                                                         but focusses only on major crops for which the
represent the output of the sector as a whole, it                                                                  data are more reliable. The main difference

Figure A.7.1: TFP Estimates from different methodologies and data sources
        200

        180

        160

        140

        120

        100

         80

         60

         40

         20

          0
              1980
                     1981
                            1982
                                   1983
                                          1984
                                                 1985
                                                        1986
                                                               1987
                                                                      1988
                                                                             1989
                                                                                    1990
                                                                                           1991
                                                                                                  1992
                                                                                                         1993
                                                                                                                1994
                                                                                                                       1995
                                                                                                                              1996
                                                                                                                                     1997
                                                                                                                                            1998
                                                                                                                                                   1999
                                                                                                                                                          2000
                                                                                                                                                                 2001
                                                                                                                                                                        2002
                                                                                                                                                                               2003
                                                                                                                                                                                      2004
                                                                                                                                                                                             2005
                                                                                                                                                                                                    2006
                                                                                                                                                                                                           2007
                                                                                                                                                                                                                  2008
                                                                                                                                                                                                                         2009




           DEA (using FAO database)                              Fuglie (2012), raw series                               Bosworth and Collins (updated series)                                             Rada 2013
Source:  Authors




  212
                                                    ﻿Annex 7: TFP: Methodology and Crop-Specific Estimates




is that it does not include horticultural             Finally, access to a unique household-level
crops (fruits and vegetables) and livestock           panel dataset allows deeper and more precise
products, and hence provides insight into             insights into the changes occurring at the farm
the performance of the more traditional crop          level. Detailed information on agricultural
sector. Nevertheless these crops cover over half      production for 1999 and 2007 for 4,000
the estimated value of production (aggregated         households from the NCAER-REDS data are
at the national level) and account for nearly         used to estimate productivity changes and
75 percent of the area under cultivation. So          various measures of efficiency. The data also
while diversification is an important strategy to     allow alternative methodologies to be used to
improve incomes and productivity in a broader         triangulate potential shortcomings associated
sense, the narrower estimates provide an              with particular methodologies. Using the
important contrast to how the majority of the         nonparametric DEA methodology, Nin-Pratt and
sector resources are being used. An important         Gautam (2013) provide a detailed breakdown of
methodological difference between this study          productivity changes at the farm level as well as
and the estimates by Rada is that it uses a more      estimates of profits lost due to different types of
detailed and comprehensive accounting for             inefficiency at the farm level. Complementing
inputs by using the estimates from the survey-        this work, Gautam, Pradhan, and Nagarajan
based annual CoC data compiled for the major          (2013) use econometric methods to estimate the
crops by the Ministry of Agriculture. These data      determinants of both technical and economic
not only allow a more complete accounting             efficiency across household status levels and
for the inputs used by farmers and but permit         determinants of technical as well as allocative
more accurate estimation of their cost shares,        efficiency, providing insights into the overall
allowing a more precise measurement of TFP.           economic efficiency at the household level.




                                                                                                    213
Republic of India: Accelerating Agricultural Productivity Growth




Table A.7.1:  Major crops included in estimates of TFP for traditional crops
Crop                                                         State
Cereals
Paddy                 Andhra Pradesh, Assam, Bihar, Haryana, Karnataka, Madhya Pradesh, Odisha, Punjab,
                      Tamil Nadu, Uttar Pradesh, West Bengal
Jowar                 Andhra Pradesh, Karnataka, Madhya Pradesh, Maharashtra, Rajasthan, Tamil Nadu
Bajra                 Gujarat, Haryana, Maharashtra, Rajasthan, Tamil Nadu, Uttar Pradesh
Maize                 Andhra Pradesh, Bihar, Himachal Pradesh, Madhya Pradesh, Rajasthan, Uttar Pradesh
Wheat                 Bihar, Haryana, Himachal Pradesh, Madhya Pradesh, Punjab, Rajasthan, Uttar
                      Pradesh, West Bengal
Pulses
Gram                  Haryana, Madhya Pradesh, Rajasthan, Uttar Pradesh
Moong                 Andhra Pradesh, Odisha, Rajasthan
Arhar                 Gujarat, Karnataka, Madhya Pradesh, Uttar Pradesh
Urad                  Andhra Pradesh, Madhya Pradesh, Maharashtra, Odisha, Rajasthan, Tamil Nadu,
                      Uttar Pradesh
Edible oilseeds	
Rapeseed & mustard    Assam, Haryana, Punjab, Rajasthan, Uttar Pradesh
Groundnut             Andhra Pradesh, Gujarat, Karnataka, Odisha, Tamil Nadu
Soybean               Madhya Pradesh
Fibre crops
Cotton                Andhra Pradesh, Gujarat, Haryana, Karnataka, Madhya Pradesh, Maharashtra,
                      Punjab, Tamil Nadu
Jute                  Assam, Bihar, Odisha, West Bengal
Other cash crops
Sugarcane             Andhra Pradesh, Bihar, Haryana, Karnataka, Maharashtra, Tamil Nadu, Uttar Pradesh
Source:  Authors




  214
                                                                                                                                         ﻿Annex 7: TFP: Methodology and Crop-Specific Estimates




Figure A.7.2: Crop-specific TFP estimates by state

                                                                 Paddy                                                                                                                               Wheat
    3.00                                                                                                                                    2.50
    2.50                                                                                                                                    2.00
    2.00                                                                                                                                    1.50
    1.50                                                                                                                                    1.00
    1.00
                                                                                                                                            0.50
    0.50
                                                                                                                                            0.00
    0.00




                                                                                                                                                   MP

                                                                                                                                                                 W. Bengal


                                                                                                                                                                              UP

                                                                                                                                                                                              Bihar


                                                                                                                                                                                                                   Haryana

                                                                                                                                                                                                                                   Punjab


                                                                                                                                                                                                                                                          Rajasthan


                                                                                                                                                                                                                                                                                 HP


                                                                                                                                                                                                                                                                                                 Gujarat
           Haryana
                        Assam

                                  T. Nadu


                                                     Karnataka
                                                                 UP

                                                                             Punjab

                                                                                      Odisha

                                                                                                      W. Bengal

                                                                                                                    Bihar

                                                                                                                             AP

                                                                                                                                   MP
   -0.50                                                                                                                                   -0.50
   -1.00                                                                                                                                   -1.00
   -1.50                                                                                                                                   -1.50

                                       1981-95                                    1995-08                                                                                                     1981-95                            1995-08

                                                                 Maize                                                                         6                                                Oilseeds
    6.00                                                                                                                                       5
    5.00                                                                                                                                       4
    4.00                                                                                                                                       3
    3.00                                                                                                                                       2
    2.00                                                                                                                                       1
    1.00                                                                                                                                       0
                                                                                                                                                    Groundnut
                                                                                                                                                                Groundnut


                                                                                                                                                                             Groundnut

                                                                                                                                                                                              Groundnut

                                                                                                                                                                                                            Groundnut

                                                                                                                                                                                                                             Groundnut


                                                                                                                                                                                                                                                     Soyabean

                                                                                                                                                                                                                                                                      Soyabean

                                                                                                                                                                                                                                                                                 Sunﬂower

                                                                                                                                                                                                                                                                                                 Sunﬂower
                                                                                                                                              -1
    0.00
           MP                    Rajasthan                         HP                 UP                            AP           Bihar        -2
   -1.00
   -2.00
                                      1981-95                                 1995-08                                                              KT             OR         TN               GJ AP                          MH MP                                     RJ         KT             MH
                                                                                                                                                                                              1981-95                         1995-08

                                                                 Cotton                                                                                                                       Sugarcane
  14.00                                                                                                                                    0.50
  12.00
  10.00                                                                                                                                    0.00
    8.00


                                                                                                                                                                                                                                                                      UP
                                                                                                                                           -0.50
    6.00
                                                                                                                                                    Ap




                                                                                                                                                                                                                                                                                       Haryana
    4.00                                                                                                                                   -1.00
                                                                                                                                                                                                                                         Karnataka
                                                                                                                                                                                                          Tamil Nadu
                                                                                                                                                                                Maharashtra




    2.00
    0.00                                                                                                                                   -1.50
           Tamil Nadu


                            Haryana

                                            Punjab


                                                                 Rajasthan

                                                                             Gujrat

                                                                                        Maharashtra

                                                                                                        Karnataka


                                                                                                                            AP

                                                                                                                                   MP




   -2.00
   -4.00                                                                                                                                   -2.00
                                                                                                                                           -2.50

                                                                                                                                                                                   1995-08                                          1981-08
                                             1981-95                                  1995-08
Source:  Kumar, Gautam and Joshi 2013.




                                                                                                                                                                                                                                                                                                    215
                      Annex 8: Changes in Farm
annexes               Revenues and Efficiency by
                      Land Size



 Figure A.8.1: Gross revenues per acre and by farm size, 1982, 1999, 2007

                             15000




                             10000




                               5000




                                  0
                                       0            10           20   30          40             50

                                             Revenue per acre 1982    Revenue per acre 1999
                                             Revenue per acre 2006

 Source:  Authors, using ICRISAT-NCAP district database.


 Figure A.8.2: Technical efficiency and farm size, 1982, 1999, 2007

                                  .8




                                 .75




                                  .7




                                 .65
                                       0            10          20    30         40           50
                                           Technical Efﬁciency 1982   Technical Efﬁciency 1999
                                           Technicl Efﬁciency 2007
 Source:  Authors, using ICRISAT-NCAP district database.



   216
                   Annex 9: Progress and
annexes            Potential for Gains through
                   Technology Capital



 Potential and Attainable Yield                        Weather data of a representative location in
 simulation model                                      each of these subzones was obtained from
                                                       the India Meteorology Department. Wherever
 The InfoCrop model (Aggarwal et al. 2006)             such climatic data was not available, data of an
 was used for simulating potential yields. This        appropriate location adjacent to the selected
 generic dynamic crop simulation model is              zones was utilized. Representative soil profile
 sensitive to variety, agronomic management,           details for each zone were input in the model
 soil, weather, flooding, frost, and pests. The        for calculation of simulated rainfed potential
 model simulates all major processes of crop           yields.
 growth, soil water and nutrient balances,
 greenhouse gas emissions, and crop-pest               Wheat simulations were done for three sowing
 interactions. It is used for estimates of potential   dates (normal, late, and very late) in line with
 yields and yield gaps, impact assessment              farmers’ practices in different regions. For
 of climatic variability and climate change,           each sowing date, coefficient of a standard
 optimizing management (planting dates,                variety, common for that time of sowing, was
 variety, irrigation, and nitrogen fertilizer),        used. For paddy, it was assumed that the crops
 genotype by environment by management                 were transplanted with the standard onset
 by pest interactions, yield forecasting, and          date of the rainy season. It was assumed that
 assessments of yield losses due to pests. The         both rice and wheat did not suffer from any
 model has been calibrated and validated in            deficiency of water or nutrients and that pests
 typical rainfed and irrigated crop growing areas.     and diseases were controlled. Simulations were
                                                       done using 10–20 years of weather data for
 A methodology was developed for estimating            each location depending upon its availability.
 regional crop production using a crop                 The average of the yields of all years was used
 simulation model together with databases for          as the simulated potential yield for a location.
 soil, management, and weather. The Planning           Simulated potential yield was aggregated for
 Commission of the Government of India has             each zone considering the weighted distribution
 divided the country into 15 agro-climatic             of planting times (in the case of wheat).
 zones and 95 subzones. The latter were used
 as the primary simulation units. Only those           The results from the models for wheat and
 subzones where rice or wheat cultivation is           rice are given in Figures A.9.1 and A.9.2. The
 significant were considered for this analysis.        figures present simulated potential yields



                                                                                                    217
Republic of India: Accelerating Agricultural Productivity Growth




(assuming no constraints to inputs), attainable                                                                             level, which is assessed to be a more realistic
yields (given the current level of irrigation), and                                                                         assessment of the current yield potential for
actual yields (average of 2006–10). The states are                                                                          each state, conditional on the level of irrigation
shown in decreasing order of attainable yield                                                                               development in the state.

Figure A.9.1: Current yield gaps for wheat by state

                               9                                                                                          Wheat
                               8
                               7
              Thousand kg/ha




                               6
                               5
                               4
                               3
                               2
                               1
                               0
                                      Punjab



                                                        Haryana



                                                                        UP



                                                                                       Rajasthan



                                                                                                    Bihar



                                                                                                                     MP



                                                                                                                                  Gujarat



                                                                                                                                                    HP



                                                                                                                                                                             West Bengal



                                                                                                                                                                                                      Maharashtra



                                                                                                                                                                                                                              Karnataka



                                                                                                                                                                                                                                              Assam
                                                                  Actual Yields (2006-10)                            Potential Yield                        Attainable Yield


Source:  Authors


Figure A.9.2: Current yield gaps for paddy/rice by state

                                                                                                                       Paddy
                               12

                               10
         Thousand kg/ha




                               8

                               6

                               4

                               2

                               0
                                    Haryana


                                               Punjab


                                                                  AP


                                                                          Tamil Nadu


                                                                                       Karnataka


                                                                                                   UP


                                                                                                            Kerala


                                                                                                                          Bihar


                                                                                                                                     West Bengal


                                                                                                                                                   Odisha


                                                                                                                                                              Chhattisgarh


                                                                                                                                                                                            Gujarat


                                                                                                                                                                                                                Maharashtra


                                                                                                                                                                                                                               MP


                                                                                                                                                                                                                                          Jharkand




                                                                       Actul Yields (2006-10)                              Attainable Yield                                                Potential Yield


Source:  Authors




  218
                                           ﻿Annex 9: Progress and Potential for Gains through Technology Capital




Figure A.9.3: Yields of released varieties: Rainfed rice, 1965–2010

            12.00
                                      Actual Yields (Combined)                       1965-80
                                      1980-95                                        1995-2010

            10.00


                           y = 0.1632 x - 317.62               y = 0.0077x +21.003                   y = 0.0058x -5.7811
            8.00



            6.00



            4.00



            2.00
                                                                                                 y = 0.0301x-58.29

            0.00
                    1960     1965    1970     1975      1980       1985    1990       1995   2000       2005     2010      2015


Source:  Authors, using DRR database.


Figure A.9.4: Yields of released varieties: Irrigated rice, 1965–2010

           12.00                                   Actual Yields (Combined)            1965-80
                                                   1980-95                             1995-2010


           10.00


                           y = 0.0968x - 185.01                  y = 0.1667x -324.82                    y = 0.0061x - 4.6864
           8.00



           6.00



           4.00



           2.00

                                                                                                  y = 0.0301x-58.29

           0.00
                    1960    1965    1970     1975       1980      1985    1990       1995    2000      2005     2010       2015


Source:  Authors, using DRR database.




                                                                                                                                  219
Republic of India: Accelerating Agricultural Productivity Growth




Figure A.9.5: Yields of released varieties: Rainfed wheat, 1965–2010

                                                Actual Yields (Combined)            1965-80
         8.00                                   1980-95                             1995-2010


         7.00            y = 0.0534x - 101.86                                                  y = 0.0355x - 66.478
                                                           y = 0.1499x - 293.84


         6.00


         5.00


         4.00


         3.00


         2.00


         1.00                                                                               y = 0.0464x - 90.156


         0.00
                  1955           1965             1975            1985            1995           2005               2015



Source:  Authors, using DWR database.

Figure A.9.6: Yields of released varieties: Irrigated wheat, 1965–2010


                                                  Actual Yields (Combined)               1965-80
                                                  1980-95                                1995-2010
          8.00
                         y = 0.015x - 24.664              y = 0.0159x - 25.887               y = 0.0541x - 102.41
          7.00

          6.00

          5.00

          4.00

          3.00

          2.00                                                                               y = 0.0464x -90.156

          1.00

          0.00
                  1955           1965             1975            1985            1995               2005           2015


Source:  Authors, using DWR database.




  220
                                        ﻿Annex 9: Progress and Potential for Gains through Technology Capital




Figure A.9.7: Wheat varieties released by decade and zone


          250



          200



          150



          100



           50



            0
                   1965-1970            1971-1980          1981-1990               1991-2000            2000-2011

                                   NWP              NEP          CZ           PZ            NHZ



Source:  Authors, using DWR database.



Figure A.9.8: Rice varieties released by decade and ecosystem

          250


          200


          150


          100


           50


            0
                    1965-1970           1971-1980          1981-1990               1991-2000            2001-2010
                                IRE         IRME          IRM          RUP           RSL          DW
                                SDW         HRUP          HRIR         IRSA          BORO         SCR


Source:  Authors, using DRR database.




                                                                                                                    221
Republic of India: Accelerating Agricultural Productivity Growth




Table A.9.1:  Farm yields, potential yields, and yield gaps in global rice mega-environments
                                                Estimated yield (t/ha) and gap (%),          Rate of change (%) relative to
                                                        in 2009 or 2010                             current values
RME                     Region                     FY             PY           Gap            FY            PY           Gapb
1          C. Luzon, the Philippines                3.9           7.0            79            0.6           0.7           0.1
           (wet season)
1          C. Luzon, the Philippines                4.6           9.5           107            0.6           0.7           0.1
           (dry season)
1.         The Philippines                          3.7           5.6            51            1.6           0.7         −0.9
1.         Indonesia                                5.0           6.5            30            0.7           1.3           0.6
1          Southern Vietnam                         5.2            na            na            1.9           1.0         −0.9
2          Jiangsu, China                           8.0          11.0            38            0.7           1.2           0.5
2          Punjab, India                            6.0          10.5            75            1.1           0.3         −0.8
3          Japan                                    6.6          11.8            80            0.2           0.7           0.3
3          Rio Grande do Sul, Brazil                7.0          10.5            50            2.0           0.7         −1.3
4          Egypt                                   10.0         12.0   c
                                                                                 25            1.1          0.6  d
                                                                                                                         −0.5
5          Madhya Pradesh, India      b
                                                    1.8           4.5           150            1.2           na            na
5          North-East Thailand    b
                                                    2.1           5.0           140            1.6           na            na
6          Central Brazil   b
                                                    2.0           3.6            80            2.2           0.7         −1.5
           Average (n = 13)                         5.2                          76          1.19          0.78         −0.39
Source:  Fischer, Byerlee, and Edmeades 2013.
Note:  
       RME refers to Rice mega-environment classified by Fischer, Byerlee, and Edmeades (2013) based on climate and
       hydro-morphology.
          a.	 All PY and FY slopes are significant at P < 0.10 or better. Note all rates of FY progress and gap closing contain the
              direct effect of CO2 rise (~0.2% p.a., see Section 2.4 of the source text.
          b.	 PYw was estimated for these rainfed cropping regions which commonly experience water shortage.
          c.	 Calculated as the difference between the rates of increase in PY and in FY (see Section 2.3 of the source text).
          d.	 Based on on-farm demonstration yields; includes contributions of management and breeding progress, and
              ignoring yield stagnation since 2006.




    222
                                         ﻿Annex 9: Progress and Potential for Gains through Technology Capital




Table A.9.2:  Wheat farm yields, potential yields, and yield gaps
                                               Predicted yield (t/ha) and gap (%),         Rate of change (%) relative to
                                                       in 2009 or 2010                            current values
WME                      Regiona                   FY            PY           Gap            FY            PY          Gap
1            Yaqui Valley, Mexico                   6.4           9.0           41           0.9g          0.3        −0.6
1            Punjab, India                          4.5           7.0           56           0.7g          0.4        −0.3
1            Jiangsu, China                         4.6           7.5           63            0.8          0.7        −0.1
4            Western Australia      b
                                                    1.8           2.6           44           1.0   g
                                                                                                           0.5        −0.5
6            Saskatchewan , Canada
                              b
                                                    2.3           3.8           69            0.8          0.6        −0.2
6            Saskatchewan , , Canada
                              b c
                                                    2.2           3.6           64            0.7          0.5        −0.2
6            North Dakota , USA
                             b
                                                    2.5           4.0           60            1.0          0.7        −0.3
6            Finland                                3.7           4.8           30            1.0          0.8        −0.2
10           Shandong/Henan, China                  5.8           8.8           52            1.7          0.7        −1.0
11           United Kingdom                         8.0         10.7            34            0.4          0.6        +0.2
             Northern France                        8.6         10.8            26            0.3          1.1        +0.8
12           Kansas , USA
                     b
                                                    2.8           3.8           36            0.7          0.4        −0.3
             Average (n = 12)                     4.43            na            48          0.83          0.61       −0.23
Source:  Fischer, Byerlee, and Edmeades 2013.
Note:  
       WME refers to wheat mega-environment classified by Fischer, Byerlee, and Edmeades (2013) based on growing
       season, temperature, moisture, and latitude.
      a.	 These are rainfed cropping regions commonly with water shortage so PYw was estimated
      b.	 Durum wheat.
      c.	 Note all rates of FY progress and gap closing contain the direct effect of CO2 rise (~0.2% p.a., see Section 2.4 of
          the source text).
      d.	 All FY and PY slopes are statistically significant at P < 0.10 or better, except for the FY slope for Northern France
          (P = 0.13).
      e.	 Calculated as FY rate of change less PY rate of change.
      f.	 The FY rates of change include small but significant weather trends (see text) for which no correction is applied
          here; two were unfavorable and one favorable.




                                                                                                                         223
Republic of India: Accelerating Agricultural Productivity Growth




Box A.9.1: The National Agricultural Innovation Project: Supporting change in India’s
national agricultural research system

The ongoing World Bank–supported National Agricultural Innovation Project (NAIP), implemented by ICAR
since 2006, is piloting a number of change initiatives in the Indian national agricultural research system.
The initiatives have the potential to catalyze system-wide efficiency, effectiveness, and productivity, and
they include:
    Competitive consortia-based funding for agricultural research to introduce pluralism in the
      national agricultural research system. A total of 188 subprojects have been approved under
      NAIP, involving 844 participating institutions. More than 90 public-private partnerships have
      been established among 174 private organizations and NGOs. This is the first time that private
      organizations from outside the ICAR and state agricultural university system have been involved as
      partners in carrying out publicly funded agricultural research in the country.
    Support for system-wide changes to improve efficiency and productivity. Particularly noteworthy
      subprojects seek to establish a financial management/management information system that would
      connect all ICAR institutions on a real-time basis; establish a National Agricultural Bio-informatics
      Grid that would provide access for scientists to high-performance computing facilities for research
      related to biotechnology; and an online examination system for recruitment of agricultural scientists
      across the country.
    Business Planning and Development Units (BPDU) piloted in 10 selected state agricultural
      universities and ICAR institutions. These units have encouraged, nurtured, and supported
      technologists and scientists to develop their research products into sound commercial ventures.
      About 31 technologies have already been commercialized through this route, generating revenues
      surpassing Rs. 10 crores in the process.
    Stronger information communication and dissemination systems to provide online
      information to agricultural scientists. These systems include an online e-publishing system for
      ICAR research journals; operationalization of the Consortium for e-Resources in Agriculture (CeRA)
      to provide online access to about 3,000 scholarly journals for 142 institutions in the national
      agricultural research system; development of a knowledge management platform (Agropedia) to
      aggregate and disseminate information; a knowledge management portal providing complete
      information on rice; a group catalog (AgriCat, http://www.agricat.worldcat.org) of 12 major libraries
      for online access by researchers and students; a new platform (KVKnet http://agropedialabs.iitk. ac.in/
      extension) and knowledge network (vKVK http://www.vkvk.in) for Krishi Vigyan Kendra (Agriculture
      Science Centre) scientists; e-courses for five Bachelor’s degree programs and strengthening of
      statistical computing in the national agricultural research system (http://www.iasri.res.in/sscnars).
Encouraged by the initial success, ICAR plans to further scale up and mainstream initiatives such as
competitive research funding and BPDUs during the 12th Plan (2012–17) using its own funds.
Source: Authors




 224
                                    ﻿Annex 9: Progress and Potential for Gains through Technology Capital




Box A.9.2: The Agricultural Technology Management Agency Model of extension

The Agricultural Technology Management Agency (ATMA) is a quasi-governmental organization registered
under the Societies Registration Act of 1860. It operates under the direction and guidance of a Governing
Board (GB) that determines program priorities, allocates funds, and assesses program impact. The
composition of the GB provides a balance between the heads of the line departments and research units
within the district and the stakeholder representatives—farmers, women, disadvantaged groups, and
private firms within the district. The GB, chaired by the District Magistrate/Collector, reviews and approves
the Strategic Research and Extension Plan and annual work plans, and sets policies and procedures for
ATMA operations.
ATMA Management Committee (AMC) serves as the Secretariat of the GB and is responsible for coordinating
and integrating extension and research activities within the district. The Project Director (PD), ATMA chairs
the AMC, which includes the district heads of line departments, zonal research station, Krishi Vigyan
Kendra (KVK, Agriculture Science Centre), NGO representatives, and two representatives from farmers’
organizations.
ATMA Personnel include the PD, supported by a Deputy PD and other staff. The PD and DPD are taken on
deputation, and the support staff is hired on a contract basis. To facilitate research and extension linkages
within the district, if the PD is from the research system, then the DPD is from the extension system and
vice-versa. Actual implementation of ATMA activities is done by the line departments.
The Block Technology Team (BTT) consists of block-level line department (agriculture, horticulture, animal
husbandry, dairy, fisheries, forestry, and sericulture) officers and subject matter specialists, and it is headed
by a Block Technology Manager. A designated scientist from the local KVK also attends the BTT meetings
and provides a link with the agricultural research system. The BTT consults with the Farmer Advisory
Committee (FAC) and develops a comprehensive extension program called the Block Action Plan (BAP),
consistent with farmers’ needs.
The Farmer Advisory Committee (FAC) is composed of farmers who represent different disciplines
(agriculture, horticulture, livestock, and so on) and socioeconomic groups. Following the 2010 revised ATMA
guidelines, FACs are now established at State, District, and Block levels; previously there was only a block
FAC (BFAC). The BFAC advises the BTT on extension priorities for the block. It also reviews and approves the
annual BAP prepared by the BTT before its submission to ATMA for funding. Then, the BFAC monitors and
provides feedback to BTT on BAP implementation.
The Farm Information and Advisory Center (FIAC) is a block-level facility which includes an office for the
BTT convener, a meeting room, and office space for the operator of the FIAC computer, with internet
connectivity. It has become the single-window delivery mechanism for extension programs within the
block. The internet access serves as an important information resource for all participants.
Farmer Organizations (FOs) are a key element of the group approach and focus on diversification into high-
value crops and products. The farmers are organized into farmer interest groups around specific crops or
products for which there is a market demand and that are appropriate for the agro-ecological conditions
and resources of each group. To successfully supply different markets, achieve economies of scale, and
create an efficient supply chain, these groups are organized along crop or product lines as block- and
district-level farmer federations.
Source: Authors




                                                                                                            225
Republic of India: Accelerating Agricultural Productivity Growth




Box A.9.3: Changes introduced in ATMA through the 2010 revised guidelines

    Provides for dedicated manpower at various levels. In 2005 when the pilot phase was scaled up, the
      scheme did not provide for dedicated manpower support at the different organizational levels. Work
      pertaining to ATMA was mostly looked after by staff of line departments in addition to their other
      duties, leading to the neglect of ATMA work. The revised guidelines provide for a State Coordinator,
      faculty, and support staff at the State Agricultural Management and Training Institute (SAMETI);
      a Project Director, Deputy Project Director, and support staff at the district level; and a Block
      Technology Manager and Subject Matter Specialist at the block level.
    During Phase II of the program, the extension system below the block level was weak, with
      inadequate links to the levels above. The revised guidelines seek to address this weakness by
      providing for a “farmer friend” for every two villages.
    Additional activities have been added to the “ATMA cafeteria” (the list of extension activities that can
      be funded), such as farmer schools. Unit costs have been enhanced for some of the activities.
    Farmers Advisory Committees (FACs) at the state, district, and block levels comprise farmer
      representatives. The earlier guidelines provided for FAC at the block level only.
    Support to SAMETIs for essential infrastructure.
    Delegation of powers to State Level Sanctioning Committees (SLSCs) set up under RKVY, to approve
      the State Extension Work Plan (SEWP) prepared under the Extension Reforms Scheme.
Source: Adapted from ATMA Guidelines 2010, available at http://Agricoop.nic.in/




 226
annexes
                      Annex 10: Livestock Sector:
                      Status and Performance



                Share of livestock in gross value of output and growth of the agricultural sector, TE 1992-
 Table A.10.1:  
                93 and TE 2008-09, at 2004-05 prices (%)

                           Livestock share in      Annual livestock      Annual agric. sector     Livestock share in
                            agricultural VOP        sector growth             growth                 agric. growth
  State                    1992-93    2008-09     1992-93     2008-09     1992-93     2008-09    1992-93      2008-09
  Haryana                    27.8       32.0         3.6         3.8         2.9         2.9       34.9        41.1
  Punjab                     26.2       33.1         4.1         2.5         2.0         1.7       57.8        47.2
  Rajasthan                  33.9       38.5         5.2         1.7         3.3         2.3       53.2        29.3
  Himachal Pradesh           27.3       28.3         3.7         3.3         3.3         3.5       30.3        26.2
  Gujarat                    19.6       25.4         5.3         6.2         3.4         5.0       30.0        31.4
  Uttar Pradesh              21.0       28.2         4.5         3.6         3.0         1.5       31.8        69.7
  Madhya Pradesh             26.8       27.0         3.1         3.2         3.3         2.9       25.5        29.9
  Tamil Nadu                 27.4       29.0         1.9         3.5         2.4         1.8       21.6        57.4
  Maharashtra                22.1       20.1         4.0         3.1         3.9         4.1       22.9        15.0
  Andhra Pradesh             22.5       30.2         6.3         5.3         3.0         4.6       46.7        35.4
  Bihar                      28.7       36.8         6.2         7.4         6.6         3.9       27.3        69.1
  Kerala                     19.5       21.0         2.9         0.7         2.5         0.6       23.2        25.8
  West Bengal                21.6       20.9         2.5         2.1         3.5         1.5       15.4        29.4
  Assam                      10.6       13.0         1.2         3.5         1.9         0.1        6.7          -
  Odisha                     10.2       18.6         4.2         7.7         0.4         3.4      104.3        42.3
  Northeastern states        22.3       23.1         4.5         4.5         3.9         2.7       25.5        38.3
  India                      23.3       26.8         3.9         3.6         2.9         2.7       31.1        36.3
 Source:  Birthal and Negi 2012.
 Note:  Bihar includes Jharkhand; Madhya Pradesh includes Chhattisgarh; Northeastern states excludes Assam;
 Uttar Pradesh includes Uttarakhand.




                                                                                                                 227
Republic of India: Accelerating Agricultural Productivity Growth




Table A.10.2:  Livestock population, output, and yield, 1990, 2000, and 2010
                                                 Year                                        % change
                                                                             2000 over       2010 over     2010 over
                                 1990           2000            2010           1990            2000          1990
Milk output, million MT
Cow milk                     22.2                  33           54.9            48.6               66.4     147.3
Buffalo milk                 29.1                43.4           62.4            49.1               43.8     114.4
Goat milk                      2.4                3.3            4.6            37.5               39.4      91.7
Total                        53.7                79.7          121.9            48.4               52.9     127.0
Milk yield, kg/in-milk animal/year
Cow                           732               1003            1284              37                28       75.4
Buffalo                     1122                1479            1679             31.5              13.5      49.6
Goat                          100                122             151              22               23.8      51.0
Population , million head
Cattle                     202.5                191.9           210.2             -5.2               9.5       3.8
Buffaloes                    80.6                93.8           111.3            16.4               18.7      38.1
Dairy cows                   30.4                32.9            42.8              8.2               30       40.7
Dairy buffaloes              25.9                29.7            37.1            14.8               24.9      43.4
Goats                      113.2                123.5            154               9.1              24.7      36.0
Dairy goats                  23.8                26.7            30.5            12.2               14.2      28.2
Sheep                        48.7                59.4             74               22               24.6       52
Pigs                         11.9                13.4             9.6            12.6              -28.4     -19.4
Chickens                      268                374            773.8            39.6             106.8     188.7
Ducks                        22.6                30.4             26             34.5              -14.5      15.0
Slaughter animals, million head
Cattle                       10.3                 9.6            10.6             -6.8             10.4       2.9
Buffalo                        7.8                9.1            10.8            16.7              12.5      38.5
Goat                         43.0                46.9            58.7              9.1             25.2      36.5
Sheep                        15.1                18.4            24.1            21.8              31.0      59.6
Meat and eggs, million MT
Cattle meat                 1.036               0.981           1.087             -5.4             10.8        4.9
Buffalo meat                1.078               1.256           1.489            16.5              18.6       38.1
Total                       2.114               2.237           2.576              5.8             15.2       21.9
Goat meat                    0.43               0.469           0.587              9.1             25.2       36.5
Sheep meat                  0.181               0.221           0.289            22.1              30.8       59.7
Pig meat                    0.413               0.466           0.333            12.8              28.6      -19.4
Chicken meat                0.362               0.864           2.193           138.7             153.8     505.8
Duck meat                    0.03                0.04           0.038            33.3                -5       26.7
Hen eggs                    1.161               2.035           3.378            75.3               66      191.0
Source:  Estimated from http://faostat.fao.org/site/603/default.aspx#ancor accessed 22/11/2012.




  228
                                                                                                             ﻿Annex 10: Livestock Sector: Status and Performance




                griculture (crops) and livestock GDP annual growth rates
Figure A.10.1: A


        20.0%

        15.0%

        10.0%

         5.0%

         0.0%
                 1952
                 1954
                        1956
                               1958
                                      1960
                                             1962
                                                    1964
                                                           1966
                                                                  1968
                                                                         1970
                                                                                1972
                                                                                       1974
                                                                                              1976
                                                                                                     1978
                                                                                                            1980
                                                                                                                   1982
                                                                                                                          1984
                                                                                                                                 1986
                                                                                                                                        1988
                                                                                                                                               1990
                                                                                                                                                      1992
                                                                                                                                                             1994
                                                                                                                                                                    1996
                                                                                                                                                                           1998
                                                                                                                                                                                  2000
                                                                                                                                                                                         2002
                                                                                                                                                                                                2004
                                                                                                                                                                                                       2006
                                                                                                                                                                                                              2008
                                                                                                                                                                                                                     2010
         -5.0%

        -10.0%

        -15.0%

        -20.0%                                                                          Agriculture                                            Livestock



Source:  Authors, using CSO National Accounts data.




Table A.10.3:  Extent of contract broiler production by state in India, 2004–05
                                                       Total production                                            Production under contract                                             % production under
State                                                (million birds/month)                                           (million birds/month)                                                    contract
Tamil Nadu                                                                 18.5                                                                16.5                                                           90
Karnataka                                                                       7.4                                                             6.5                                                           87
Andhra Pradesh                                                             16.0                                                                 9.5                                                           60
Maharashtra                                                                11.0                                                                 8.5                                                           73
Subtotal                                                                   52.9                                                                41.0                                                           78
Gujarat                                                                         2.6                                                             0.9                                                           35
West Bengal                                                                14.7                                                                 3.0                                                           20
Northern states                                                            30.0                                                                 2.0                                                            7
Other states                                                               30.0                                                                 1.0                                                            3
Total                                                                    130.2                                                                 47.9                                                           37
Source:  Fairoze et al. 2006.




                                                                                                                                                                                                                            229
Republic of India: Accelerating Agricultural Productivity Growth




               Share of livestock sector in ground-level disbursement of agricultural credit, 2000–01 to
Table A.10.4:  
               2009–10, at 2004–05 prices
Year              Total agric credit (Rs billions)   Livestock credit (Rs billions)   % share of livestock
2000–01                         635                               26                          4.1
2001–02                         719                               25                          3.5
2002–03                         781                               30                          3.8
2003–04                         926                               31                          3.4
2004–05                       1,253                               31                          3.4
2005–06                       1,728                               70                          4.1
2006–07                       2,060                               72                          3.5
2007–08                       2,183                               77                          3.5
2008–09                       2,396                               83                          3.4
2009–10                       2,939                               78                          2.7
Source:  Birthal and Negi 2012.




  230
annexes
                      Annex 11: Public Expenditure
                      Patterns and Impact




 Table A.11.1:  Analysis of NFSM expenditures for 2013–14 (Rs)

  Expenditure category                                         Gujarat   Bihar       Andhra Pradesh
  Input distribution                                           2,275     3,895           8,500
  (seeds, micronutrients, other inputs)
  Subsidy for agricultural equipment                             580     1,750           1,120
  Subsidy toward irrigation equipment                            320     1,085           1,625
  Training and capacity building                                   60       0              100
  Other local initiatives                                        320        0                0
  Project management                                             110      390              160
  Total                                                        3,665     7,120          11,505
  Proportion of subsidies in the overall budget                  87%      95%              98%
 Source:  Authors, using information from http://www.nfsm.gov.in



 Table A.11.2:  Analysis of RKVY expenditure by approved projects for 2012–13 (Rs)

  Expenditure category                                         Gujarat   Bihar       Andhra Pradesh
  Input distribution                                               350     570             285
  (seeds, micronutrients, other inputs)
  Subsidy for agricultural equipment                               100     160             320
  Subsidy toward irrigation equipment                              120       0               0
  Training and capacity building                                    10       5              80
  Institutional strengthening (including support                   480     100             190
  for building public assets)
  Total                                                          1,060     835             875
  Proportion of subsidies in the overall budget                   42%      87%            69%
 Source:  Authors, using information from http://rkvy.nic.in




                                                                                                 231
Republic of India: Accelerating Agricultural Productivity Growth




                egional trends in capital expenditure, cumulative 2000–09 (2004/05 prices)
Figure A.11.1: R

                                40,000
                                35,000                                                    Irrigation                                       Infrastructure

                                30,000
         Rs. Per ha (GCA)




                                25,000
                                20,000
                                15,000
                                10,000
                                 5,000
                                      0
                                                       North                  Eastern                               Central                              Southern                                North-East                             All India

Source:  Singh 2011.


               Percent area irrigated under major crops by state, 2008–09
Figure A.11.2: 
          120.0

          100.0

                 80.0

                 60.0

                 40.0

                 20.0

                            0.0
                                                                                                                                                                               Meghalaya
                                   Punjab
                                            Haryana
                                                      UP
                                                           Bihar
                                                                   TN
                                                                        WB
                                                                             AP
                                                                                  UTTK.
                                                                                          Gujarat
                                                                                                    Tripura
                                                                                                               Odisha
                                                                                                                        Rajasthan
                                                                                                                                    MP
                                                                                                                                         Karnataka
                                                                                                                                                     Chhat.
                                                                                                                                                              Goa
                                                                                                                                                                    Manipur


                                                                                                                                                                                           Nagaland
                                                                                                                                                                                                      HP
                                                                                                                                                                                                            Mahar.
                                                                                                                                                                                                                     Kerala
                                                                                                                                                                                                                              Mizoram
                                                                                                                                                                                                                                         Jhark.
                                                                                                                                                                                                                                                  Sikkim
Source:  Authors, using DES, MOA data.                                                                                                                                                                                                                     Assam


               Impact of different public expenditures
Figure A.11.3: 
          25.0
                                                                                          1960s - 1970s                                  1980s                                1990s

          20.0


          15.0


          10.0


                5.0


                            0
                                      Roads                 Education         Irrigation                      Irrigation                  Fertilizer                     Power                              Credit                      Agricultural
                                                                             investment                       Subsidies                   Subsidies                     Subsidies                          Subsidies                       R&D
Source:  Fan, Gulati, and Thorat 2008.


  232
                                                                                                                     ﻿Annex 11: Public Expenditure Patterns and Impact




               Percent area irrigated by crop, all India 2008–09
Figure A.11.4: 

          100
           90
           80
           70
           60
           50
           40
           30
           20
           10
           0
                Wheat



                                 Rice

                                        Maize

                                                Coarse Cereals

                                                                 Bajra

                                                                         Jowar

                                                                                 Total Cereals



                                                                                                 Cram

                                                                                                        Arhar

                                                                                                                Total Pulses



                                                                                                                               Rapeseed & Mustard

                                                                                                                                                    Sunﬂower

                                                                                                                                                               Groundnut

                                                                                                                                                                           Soyabean

                                                                                                                                                                                      Oilseeds



                                                                                                                                                                                                 Sugarcane

                                                                                                                                                                                                             Tobbacco

                                                                                                                                                                                                                        Cotton
                        Barley




Source:  Authors, using DES, MOA data.




                                                                                                                                                                                                                                 233
annexes
                  Annex 12: Status of
                  Agricultural Market Reforms




 Regulatory impediments to                          handling, and marketing infrastructure, as
 marketing: ECA Act and APMC Act                    well as promoted fragmentation of markets
                                                    (through movement restrictions and cess/tax
 The two most pervasive regulations affecting       implications). Direct marketing arrangements
 the functioning of markets and trade are the       (such as CF or direct purchases by food
 Essential Commodities Act (ECA, restricting        processers from farmers) are constrained by
 movement and storage of agricultural products)     these regulations, preventing improvements in
 and the Agricultural Produce Marketing             value chain efficiencies. Thus the full potential
 Committees Act (APMC Act, under which              for food processing remains untapped, and
 agricultural marketing is restricted through       the industry remains dominated by low-level
 regulated markets with licensed traders).          processing in informal enterprises (Bhavani,
 There are also a number of other regulations,      Gulati, and Roy 2006, Morisset and Kumar
 including the Small Scale Industry Reservation     2008). Farmers as well as consumers bear the
 (under which most food processing was              brunt of these regulations, which have fostered
 reserved for small firms until 1997), and more     uncompetitive market structures at the mandi
 general policies that also affect other sectors,   (wholesale) level, depressed farmer prices, and
 such as tax policy, border and commercial          raised final prices to consumers. The specifics of
 policies, food laws, and labor policies. Many of   the two regulations are discussed in more detail
 the regulations have been relaxed over time,       below.
 allowing for a potentially more important role
 and better investment opportunities for the
                                                    The Essential Commodities Act
 private sector. Implementation of the reforms
 has been either slow or uneven, however, with      The ECA provides for the regulation and control
 reversals on several occasions.                    of production, distribution, and pricing of
                                                    commodities which are classified as “essential”
 The ECA and the APMC Act affect market             (for maintaining or increasing supplies or for
 development, efficiency, and costs in a number     assuring equitable distribution and availability
 of ways. The zoning and storage restrictions       at fair prices). Under the Act, various ministries/
 under the ECA and the requirements that all        departments of central government, and under
 produce be sold at a limited number of licensed    the delegated powers, the state governments/
 and regulated markets have created significant     union territory administrations issue orders for
 disincentives for private investment in storage,   regulating production, distribution, pricing,



  234
                                                            ﻿Annex 12: Status of Agricultural Market Reforms




Box A.12.1: LIST OF COMMODITIES DECLARED ESSENTIAL UNDER THE ESSENTIAL
COMMODITIES (AMENDMENT) ACT, 2006
No. 54 of 2006 (24th December 2006)

(1)	Drugs:
	       Explanation: For the purposes of this Schedule, “drugs” has the meaning assigned to in the clause (b) of
        section 3 of the Drugs and Cosmetics Act, 1940 (23 of 1940)
(2)	Fertilizer, whether inorganic, organic or mixed;
(3)	Foodstuffs, including edible oilseeds and oils;
(4)	Hank yarn made wholly from cotton;
(5)	    Petroleum and Petroleum Products;
(6)	    Raw jute and jute textiles;
 (7)	   (i)	Seeds of food-crops and seeds of fruits and vegetables;
  	     (ii)	Seeds of cattle fodder; and
  	     (iii)	 Jute seeds;	
  	     (iv)	Cotton seed*.
* Added vide Notification No.S.O.3267 (E) dated 22nd December, 2009.


and other aspects of trading with respect to              fertilizers (Box A.12.1).1921 There are penalties
the commodities declared as essential. The                for noncompliance. During 2006–08, state
enforcement/implementation of the provisions              and union territory governments prosecuted
of the ECA lies with the state governments and            14,541 persons under provisions of the ECA and
union territory administrations.                          secured conviction in 2,310 cases.

The ECA gives the government wide-ranging                 The inherent uncertainty in the manner in
powers to control the production, storage,                which the regulations and restrictions on the
transportation, and pricing of essential                  production, pricing, movement, and storage
commodities. Although these controls (for                 of commodities are applied creates a strong
instance, limits on the stocks that can be                disincentive for significant private investment
held by private players or restrictions on                in marketing the scheduled commodities.
the movement of commodities) are imposed                  A major casualty is investment in grain
sporadically and usually in inflationary                  storage and handling infrastructure, as well
situations, they can change suddenly in a short           as innovations in marketing, resulting in
period.                                                   inefficiency, high costs for consumers, and low
                                                          prices for producers, outcomes that are contrary
The list of scheduled commodities under the               to the Act’s original intent.
Act changes from time to time. In recent years
                                                          192	The current list includes drugs; fertilizer (inorganic, organic,
the list has been pruned significantly and                    or mixed); foodstuffs (including edible oilseeds and oils);
presently includes the seven broad categories                 hank yarn made only from cotton; petroleum and petroleum
                                                              products; raw jute and jute textiles, and seeds of food-crops,
of commodities (compared to 70 commodities                    fruits and vegetables, cattle fodder, jute, and cotton (cotton
in 1989), including foodstuffs, seeds, and                    seed was introduced in December 2010).




                                                                                                                       235
Republic of India: Accelerating Agricultural Productivity Growth




The Agricultural Produce Market                    (Umali-Deininger and Sur 2007; Fafchamps,
(Development and Regulation) Act                   Vargas-Hill, and Minten 2008).

The Agricultural Produce Market (Development       By restricting the establishment of additional
and Regulation) Act—APM(D&R)—regulates and         markets and limiting the number of commission
governs the buying and selling of agricultural     agents that can operate within the mandi,
produce through the Agriculture Produce            the net effect has been to stifle competition,
Marketing Committee (APMC) established by          generate economic rents for a small number
each state. The Act empowers states to establish   of market operators, promote inefficiency
wholesale markets for agricultural produce         in marketing, lead to under-investment in
(known as mandi or “APMC” markets). It confers     physical infrastructure by the committee,
wide powers to the APMCs to construct and          and generally create strong disincentives for
manage agricultural markets and regulate all       private investment in agricultural marketing
aspects of marketing, including the levy of a      and processing (including investment in more
user fee for transactions taking place both on     efficient markets and supply chains). Ultimately
and off the wholesale market yards.                these oligopolistic marketing arrangements
                                                   reduce the prices received by farmers and raise
The Act extends to whole of the state and makes    the prices paid by consumers. Over time, strong
these markets mandatory conduit for the trading    vested interests have developed in maintaining
of agricultural produce. The limited number        status quo (commission agents, “weigh men,”
of licenses issued by the APMCs to traders and     unions, and others) and resisting or derailing
commission agents in the established markets       reforms.
restricts the choices of sellers and buyers, who
are not allowed to conduct transactions directly   Recognizing the “malfunctioning of regulated
and outside the mandi system.                      markets” and the “need for more transparency
                                                   and accountability in the functioning of these
The management of the APMCs has changed            markets,”1932 the government initiated reforms
significantly from what was envisaged when         to facilitate the development agricultural
the regulation was put in place. Bureaucrats       markets and to facilitate greater participation
have assumed greater importance in the             of private sector through legislation, developing
management of the APMCs, replacing the             the Model State Agricultural Produce Marketing
farmer representatives who were originally         (Development and Regulation) Act, 2003 for
intended to predominate. The APMCs have            guidance and adoption by state governments.
created barriers to entry for newcomers,
resulting in limited numbers of traders and a      The Model Act provides for the establishment
large command area for each market. Market         of markets by other agencies (including the
fees have become a source of income for            private sector), direct marketing and purchase
the government instead of being reinvested         centers, CF, electronic trading, and promotion
in market infrastructure (Acharya 2004).           of public-private partnerships for managing and
                                                   developing agricultural markets in the country.
Despite large revenues from market fees, the
infrastructure in most markets is deficient, as
revenues are used for other purposes               193	GOI (2007).




 236
                                                                   ﻿Annex 12: Status of Agricultural Market Reforms




Table A.12.1:  Average size of holding of contract growers (acres) by crop and agency
                                                                                     Avg. size Avg. size of       % area
                                                                                    of contract holding in       under CF
Study and year           Contracting agency and place                   Crop          grower      state            crop
Singh (2002)            Frito Lay (Pepsi), Punjab                   Potato                53        9.5                8
-do-                    Nijjer Agro, Punjab                         Tomato                22        9.5              23
-do-                    HLL, Punjab                                 Tomato                78        9.5              33
-do-                    Frito Lay (Pepsi), Punjab                   Chilly                90        9.5              4.5
Dev and Rao (2005)      AP Govt. and various                        Oil palm              10          3              40
                        processors
-do-                    BHC Agro India, AP                          Gherkins                  7           3           15
Asokan and Singh (2006) A M Todd, Punjab                            Mint                     57         9.5
Kumar (2006)            Many multinationals and local               Many crops               37         9.5           12
                        firms in Punjab                             together
Singh (2008)            McCain Foods, Gujarat                       Potato                19           6.45          21
-do-                    Frito Lay (Pepsi) Punjab                    Potato                63            9.5          53
-do-                    A M Todd, Punjab                            Mint                  40            9.5          27
R Singh (2008)          Frito Lay, (Pepsi) Punjab                   Potato                75            9.5            -
Pritchard & Connell,    Karnataka and AP (AVT                       Chilly           22.5-42                      25-70
2011                    McCormick)
Source:  Singh 2012.



               Average size of holding of retail chain contact/contract vegetable growers by crop and
Table A.12.2:  
               agency
                                                                                                              Avg % area
Fruit and vegetable                                                Avg size of CF           Avg size of         under
chain                        Location               Crop             holding              holding in state    vegetables
ITC’s Choupal Fresh       Chandigarh          Cauliflower,              9.91               9.36 (PB) and          37
                          region (Punjab/     bottlegourd                                      5.26 (HR)
                          Haryana)
Reliance Fresh (RF)       Ahmedabad           Cauliflower,               15.9                       6.45         47
                          region              cabbage
ABRL’s More               Ahmedabad           Cauliflower,              15.43                       6.45         71
                          region              tomato
ABRL’s More               Bangalore           Cauliflower,               7.52                     4(2.9)*        65
                          region              tomato
RF/ABRL’s More            Belgaum             Cauliflower,              16.97                       4(5)*        33
(thru supplier)           region              tomato
Namdhari –Fresh           Bangalore           Okra, baby                 4.56                     4(2.9)*        64
                          region              corn
Source: Adapted from Singh 2012.
Note: * figures in brackets are average land holding size in the study district/region.




                                                                                                                      237
Republic of India: Accelerating Agricultural Productivity Growth




Table A.12.3: Status of APM (D&R) amendment and reforms

                                            Direct       Contract      Private        Single    Single-point
 Sl. No.                  State            marketing     farming       markets       license        levy
    1      Andhra Pradesh                                                                         
    2      Assam                                                                     
    3      Chhattisgarh                                                                           
    4      Gujarat                                                                               
    5      Goa                                                                                   
    6      Himachal Pradesh                                                                      
    7      Jharkhand                                                                             
    8      Karnataka                                                                             
    9      Nagaland                                                                              
   10      Madhya Pradesh                                                                        
   11      Maharashtra                                                               
   12      Mizoram                                                                               
   13      Odisha                                                       
   14      Rajasthan                                                                             
   15      Sikkim                                                                                
   16      Tripura                                                      
   17      Haryana                                          
   18      Tamil Nadu                     Act Not Amended but APM (D&R) Act provides for setting up
                                          private markets and direct marketing under section 8 - (1)–C.
   19      Uttar Pradesh                  Act not amended, provides for bulk purchases and single license
                                          through executive order.
   20      Bihar                          APMC Act is repealed w.e.f. 1.9.2006.
   21      Punjab, Meghalaya,             Act not amended
           Uttarakhand, West Bengal,
           Puducherry, Chandigarh,
           NCT of Delhi
   22      Kerala, Manipur, Andaman       States/union territories where there is no APMC Act and hence no
           and Nicobar Islands, Dadra and reforms
           Nagar Haveli, Daman and Diu,
           and Lakshadweep.
Source: Patnaik and Sharma 2013.




  238
                                                             ﻿Annex 12: Status of Agricultural Market Reforms




Table A.12.4: Status of rules notification for APM(D&R) Act amendments
Sl.No                   State                     APM (D&R) Act amended               Amended rules notified
   1    Andhra Pradesh                                                                          
   2    Assam                                                                                    -
   3    Chhattisgarh                                                                            
   4    Gujarat                                                                                  -
   5    Goa                                                                                     
   6    Himachal Pradesh                                                                        
   7    Jharkhand                                                                                -
   8    Karnataka                                                                               
   9    Nagaland                                                                        *Partially notified
  10    Madhya Pradesh                                                                          
  11    Maharashtra                                                                             
  12    Mizoram                                                                         *Partially notified
  13    Odisha                                                                                  
  14    Rajasthan                                                                               
  15    Sikkim                                                                                   -
  16    Tripura                                                                                  -
  17    Haryana                              Partially for contract farming                      
Source: Patnaik and Sharma 2013.

Table A.12.5:  Some private sector initiatives
Name of state/
union territory                    Initiatives taken                                     Outcome
Andhra Pradesh      zz Act amended on 26-10-2005                     zz Contract farming and direct purchase
                    zz Permission to NDDB for private market         zz Heritage (F&V), Reliance (F&V), More
                                                                        (F&V), Metro (F&V), Saguna Poultry Farms
                                                                        (Poultry), Sri Satyanarayana Cold Storage,
                                                                        Sri Bhuvaneshwara Multiplex Pvt.
                                                                     zz Srini Mega Food Park given special status
                                                                        for the purpose of sourcing raw material
Assam               zz Act amended on 19-01-2007
                    zz North  East Mega Food Park given special
                       status for the purpose of sourcing raw
                       material
Bihar               zz Act repealed w.e.f. 01-09-2006                zz No  progress on establishment of private
                    zz Private market near Patna has been               market reported
                       advertised and awarded (Temptation Foods)     zz Retail network established by M/S
                    zz The erstwhile markets are being                  Sambodhi in Patna City
                       operated by association of traders under      zz A few food-processing units have been
                       supervision of Officer on special duty           established
                       (mostly SDM) without regulation of            zz Kaventer Food Park has been sanctioned
                       trading practices which is creating various      by MoFPI
                       issues and problems


                                                                                                               239
Republic of India: Accelerating Agricultural Productivity Growth




Name of state/
union territory                   Initiatives taken                                   Outcome
Chandigarh        zz Partially amended                              zz No progress on establishment of private
                  zz Private market was proposed but                   market reported
                     government has not been able to attract
                     any private sector player
Chhattisgarh      zz Act amended on 10-02-2006 & 2011               zz Private market by wholesalers has been
                                                                       established at Raipur
                                                                    zz Direct sale to traders at farm-gate by
                                                                       vegetable growers
                                                                    zz Purchase of cereals by traders from farm-
                                                                       gate
                                                                    zz Contract farming of safflower by Marico
                                                                       (Saffola), of medicinal plants by CG
                                                                       Herbals
Goa               zz Act   amended on 06-08-2007

Gujarat           zz Act  amended on 01-05-2007                     zz Contract   farming by processors and
                  zz Proposal  to set up market in public-private      exporters: McCain (potatoes), Desai
                     partnership (PPP) mode in Ahmadabad,              Exports (bananas), Bodal Agro (F&V,
                     plus four terminal markets in PPP mode            pomegranates), Ambika Food Produce
                     proposed along Delhi–Mumbai freight               (F&V), Gujarat Coop. Cotton Federation
                     corridor                                          (cotton), Jayant Agro Organics, Janani
                  zz Single license                                    Industries
                  zz Anil Mega Food Park given special status       zz Farm-gate purchase by exporters
                     for the purpose of sourcing raw material
Haryana           zz Partially amended                              zz Contract   farming in basmati rice
                  zz Terminal Market project in PPP mode at
                     Gannaur is in the pipeline
                  zz E licensing has been introduced

Himachal          zz Act amended on 26-05-2005                      zz Direct purchase against single license
Pradesh                                                                under Direct Marketing License
                                                                       provisions by corporate players like Adani
                                                                       Agri Fresh, Dev Bhumi, Reliance Fresh,
                                                                       Fresh & Healthy, Mother Dairy
                                                                    zz Contract farming by processors (Himalaya
                                                                       International)
Jharkhand         zz Act amended on 16-07-2007
                  zz Jharkhand   Mega Food Park given special
                     status for the purpose of sourcing raw
                     material
Karnataka         zz Act amended on 16-08-2007                      zz Contractfarming by Global Green
                  zz Permission for private markets granted at        (gherkins), Marico (Saffola), Ugar Sugar
                     3 locations                                      works (barley), Unicorn (gherkins), Katra
                  zz Special status to the market set up by           Phyto Chem (marigold)
                     NDDB




 240
                                                          ﻿Annex 12: Status of Agricultural Market Reforms




Name of state/
union territory                Initiatives taken                                  Outcome
               zz Single license                                zz Direct purchase under Direct Marketing
               zz Integrated Mega Food Park given special          provisions by Metro Cash & Carry (F&V),
                  status for the purpose of sourcing raw           ITC, Wilchy Agro Products
                  material
Madhya Pradesh zzAct amended on 15-06-2003                      zz Contract farming done by Cargill India,
               zz Chhindwada Mega Food Park given special         Hindustan Uni Lever, ITC, Sanjivini
                  status for the purpose of sourcing raw          Orchards (pomegranates), LT Overseas
                  material                                        (basmati rice)
                                                               zz Direct purchase by cold store units,
                                                                  processors
Maharashtra        zz Act amended on 11-07-2006                zz Private markets established
                   zz MSAMB is actively setting up markets in  zz Contract Farming by Marico ( Saffola), ITC
                      PPP mode                                    & Pepsico (potatoes), More, Reilance, Big
                   zz Market infrastructure in PPP mode           Bazaar (F&V), KB Export (vegetables for
                   zz F&V trade deregulated and exempted from     export)
                      market fee
                   zz Paithan Mega Food Park given special
                      status for the purpose of sourcing raw
                      material
                   zz Direct Marketing licenses have been
                      issued to more than 80 companies/traders
Odisha             zz Act amended on, 17-05-06                 zz Contract farming by CG Herbals
                   zz Market proposed to be developed in PPP      (medicinal plants)
                      mode
                   zz MITS Mega Food Park given special status
                      for the purpose of sourcing raw material
Punjab             zz Act amended by Ordinance (lapsed after   zz Contract farming by Pepsico (potatoes),
                      six months)                                 Field Fresh (baby corn), basmati
                   zz International Fresh Farm Products Mega      exporters, Nijjar Foods (tomato), Markfed
                      Food Park given special status for the      (spinach and mustard leaves)
                      purpose of sourcing raw material
Rajasthan          zz Act amended on 18-11-2005                zz Contract farming and operations/
                                                                  procurement under Direct Marketing
                                                                  licenses by; ITC, Pepsico, Reliance Fresh
Tamil Nadu         zz No amendment required, as previous Act   zz A private modern fruit and vegetable
                      has most provisions of the Model Act        market—APPTA Market (Agricultural
                                                                  Products Producers and Traders
                                                                  Association Market)—constructed at
                                                                  Nagercoil near Kanyakumari in
                                                                  Tamil Nadu
Uttar Pradesh      zz Act not amended, but provision made vide zz Haldiram, ITC, and Pepsico are major
                      administrative order for bulk purchase      private players in direct marketing from
                                                                  farmers; Big Bazaar and Spencer’s (retail
                                                                  outlets) procure from registered vendors.
Source: Patnaik and Sharma 2013.




                                                                                                       241
annexes
                   Annex 13: Food Processing:
                   Structure and Performance




 T
       he organized manufacturing sector in           not covered by the Annual Survey of Industries
       India comprises factories registered under     (ASI). The data for both the rounds covered
       sections 2m(i), 2m(ii), and                    the National Industrial Classification (NIC)
 section 85 of the Factories Act of 1948. Under       2-digit codes 15–37 in terms of NIC 2004 codes.
 section 2m, factory means any premises               In addition, enterprises engaged in cotton
 including the precincts thereof: 2m                  ginning, cleaning, and baling (NIC 2004 code
 (i) wherein ten or more workers are working          01405) were also covered under the survey.
 or were working on any day of preceding              The survey covered the whole of the Indian
 twelve months and in any part of which a             Union; the entire study would be done at state
 manufacturing process is being carried on with       as well as national level. However, for the
 the aid of power or is ordinarily so carried         state-level analysis, the focus would be only
 on. 2m(ii) where in twenty or more workers           on 19 “major states”: Andhra Pradesh (AP),
 are working or were working on any day of            Assam (Ass), Bihar (Bih), Chhattisgarh (Chat),
 proceeding twelve months and in any part of          Gujarat (Guj), Haryana (Har), Jharkhand (Jhar),
 which a manufacturing process is being carried       Karnataka (Kar), Kerala (Ker), Maharashtra
 on without the aid of power or is ordinarily so      (Mah), Madhya Pradesh (MP), Odisha (Oris),
 carried on and does not include a Mine subject       Punjab (Punj), Rajasthan (Raj), Tamil Nadu (TN),
 to the operations of the Indian Mines Act, 1923,     Uttar Pradesh (UP), Uttaranchal (Uttr), and West
 or a railways running school. Under section 85       Bengal (WB). The survey covered the whole of
 of the Factories Act 1948, the state government      the Indian Union except (i) villages situated
 is empowered to notify any factory not covered       beyond 5 kilometers of bus route in the state of
 under the above two sections.                        Nagaland, (ii) inaccessible villages of Andaman
                                                      and Nicobar, and (iii) some first-stage units
 In the NSS framework, the unorganized                (numbering less than 0.1 percent of the total)
 manufacturing sector includes all                    where Economic Census 1998 (EC 98) could not
 manufacturing enterprises except: (i) those          be conducted. Thus the corresponding state/
 registered under section 2 m- (i) and 2 m            union territory-level estimates and the all-India
 (ii) of the Factories Act, 1948 and Bidi and Cigar   results presented are based on the areas under
 Workers (Conditions of Employment) Act,              survey coverage.
 1966, and (ii) those run by government/public
 sector enterprises. The term “unorganized            As per NSS, an Own Account Manufacturing
 manufacture” basically refers to all enterprises     Enterprise (OAME) runs without any hired



   242
                                            ﻿Annex 13: Food Processing: Structure and Performance




worker employed on a fairly regular basis and   Manufacturing Establishment (NDME). Finally,
is engaged in manufacturing and/or repairing    a Directory Manufacturing Establishment
activities. An establishment employing less     (DME) is one which has employed six or more
than six workers (household and hired workers   workers (household and hired workers taken
taken together) and engaged in manufacturing    together) and is engaged in manufacturing
activities is termed a Non-directory            activities.




                                                                                           243
                     Share of food manufacturing in total organized manufacturing (decadal averages; values in 2004–05
      Table A.13.1a: 
                     prices)




244
                                  AP      AS   BH           GJ     HY     HP    KT    KE   MP    MH    OR               PB    RJ   TN     UP   WB
      Factories        1980-90   28.46   47.84 11.76       10.03   13.69 9.58 20.56 18.61 22.96    9.5 22.07           16.87 12.5 23.11 26.71 16.27
                       1990-99   32.46   48.24 9.68         8.87   16.45 10.34 17.95 17.38 21.21 10.5 22.43            19.41 10.11 19.61 22.59 16.43
                       2000-09   39.59   51.98 11.67        9.19   12.19 13.41 18.69 20.28 17.09 11.78 30.19           20.78 9.14 16.98 17.45 19.11
      Employment       1980-90   17.49   59.94 8.56         9.65   10.94   3.6 15.84 33.31 9.58 9.99 6.03              17.16 6.46 14.34 28.76 7.42
                       1990-99    16.8   58.87 8.33          9.7   13.21 4.93 13.17 39.29 9.98 11.57 9.26              17.77 6.49 12.54 26.53 8.18
                       2000-09   18.81   58.42 18.38        8.59   11.33 8.84 14.23 47.1 12.66 14.41 14.49             22.86 7.86 10.74 25.53 11.62
      Investment       1980-90 5.78 11.67 1.97             4.64    3.93    1.3   9.38 4.59        1.33 6.17     0.66 8.72       2.56       5.11 12.1 1.99
                       1990-99 9.01 51.02 5.38             5.96    5.93 -23.22   5.59 -2.85      -0.64 7.46      6.6 11.59      0.78       10.5 7.37 7.83
                       2000-09 14.99 12.03 48.92            2.7    8.05 0.57     6.63 11.16       4.26 10.98    3.06 11.37      6.72       6.02 -77.61 10.67
      Capital stock    1980-90    7.07   32.43     1       4.71     3.6   0.84 8.03     3.33 1.42        6.53   1.26    6.37    2.73       5.64    7.28   3.03
                       1990-99    6.66   33.48 2.41        3.52    5.38    2.2 8.35     7.13 4.83        6.44   1.69   10.04    3.62       6.44    9.08   3.66
                       2000-09   11.23   24.75   6.3       2.94    6.54   4.28 9.65    11.78 7.68        8.38    2.8   15.44    4.71       6.91   16.87   6.44
      Gross value      1980-90   11.34   51.52 2.62        5.47    6.77   1.78 9.65    11.64   3.4       6.16   2.72   15.68    3.33       8.51   16.09   4.68
      added            1990-99   12.36   44.02 5.67         4.6    8.83   2.47 9.14    16.89 7.92        7.47   3.19   19.84     5.4       8.78   13.84   5.55
                       2000-09   13.72   24.24 25.69       3.74     7.5   3.93 10.72   17.61 12.13       7.29   3.29   27.84    7.67       7.75   16.93   5.96
      Source:  ASI, CSO.




                     Share of food manufacturing in total unorganized manufacturing (2000–01 and 2005–06; values in
      Table A.13.1b: 
                     2004–05 prices)
                                                                                                                                                                 Republic of India: Accelerating Agricultural Productivity Growth




                                  AP     AS   BH    CH        GJ     HY   HP     JH    KT   KE     MP    MH     OR     PB      RJ   TN       UP     UK    WB
      Enterprise      2000-01     16     31   30    15        14     16   30     21    11   12     12     19    21     12      20      9     21     20    20
                      2005-06     10     29   27       9      14     15   28     15    12   14     10     16    28     10      18      7     18     21    13
      Workers         2000-01     22     32   29    14         9     15   27     20    19   16     16     16    22     12      18   10       20     22    23
                      2005-06     22     32   27       8      10     15   24     16    26   19     12     10    28     11      15   10       19     28    16
      Investment      2000-01     31     38   44    24         9     20   41     24    18   22     14     26    39     17      21   14       26     32    22
                      2005-06     26     33   47       8       9     14   29     25    20   27     12     22    31     15      17   15       25     38    22
       Gross value    2000-01     26     23   25    15         7     15   27     19    20   17     11     19    27     12      15   12       17     23    21
       added
                      2005-06     28     30   30       8      10     17   13     16    33   18       9    21    25     15      17   13       21     30    17
      Source: NSS.
      Note: Investment or capital in unorganized refers to market value of fixed assets (own+hired).
      Table A.13.2a:  Shares of states in food manufacturing (decadal averages)
                                   AP      AS       BH             GJ         HY     HP      KT       KE           MP        MH     OR      PB      RJ          TN        UP     WB
      Factories       1980-90     24.29    3.55     0.74        5.14          2.15   0.38   5.65      4.31         2.11      8.69   2.15    6.8    2.09 14.43             6.89   4.63
                      1990-99     13.11    5.16     0.85        5.28          3.01   0.35   5.98 11.05             2.11 12.88        1.5   6.83    1.59 10.35 11.25              4.42
                      2000-09     11.38    2.02     0.75        5.72          3.83   0.81   7.28      2.27         3.92      16.4   1.57   5.33    1.57         7.28 20.38       3.67
      Employment      1980-90     10.62    3.27     1.16        6.26          2.97   0.57   8.04      2.27         3.86 18.71       1.15   5.49    1.93         8.63 16.98       3.68
                      1990-99     10.35    2.54     0.88        6.01          3.72   0.52   9.03      3.26         4.47 17.05       0.78   9.35    2.77         8.47 11.81       2.62
                      2000-09     23.68    3.49     1.33        5.14          2.69    0.2   5.31      3.42          3.9      8.74   1.66   5.81    2.02 16.33 10.61              4.53
      Investment      1980-90     12.83    5.79     1.46        6.06          3.13   0.21     5.1     9.69         3.32 12.24       1.22   5.69    1.34 11.12             15.5   4.73
                      1990-99      9.27    3.75     1.25           7.3        2.96   0.56   7.89      2.33         4.21 18.41       1.09   6.18    2.43 10.82              15    3.26
                      2000-09      9.63    4.61     1.57        7.02          2.56   0.29   6.42      2.17         5.24      18.6     1    6.22    2.22         9.67 16.38       3.41
      Capital stock   1980-90       9.2    5.06         1.3     5.78          3.29   0.21   5.83      4.38         4.73 18.88       0.78    7.9    1.99 10.26 14.31              3.24
                      1990-99     19.34   4.34      2.57        5.92          2.08   0.11   6.17      3.12          4.3      7.88   1.77   5.21    1.92 15.31 11.77              5.08
                      2000-09     11.08   6.09      2.68        5.98          2.16   0.11   5.45      7.47         3.01 11.06       0.77   4.85    1.25         10.8 19.11       5.48
      Gross value     1980-90       7.8   5.07      1.86        7.22          2.02   0.11   6.69      1.33         4.72 20.69       0.45   6.68    1.82         8.08 17.12        3.2
      added           1990-99     11.58   4.91      1.66           8.2        2.06   0.14   6.43      1.44         2.29 20.48       0.62   5.42    2.05         8.48 15.65        3.6
                     2000-09       7.46   8.41      1.98        6.38          2.47   0.14   5.76      3.95          2.3 16.47       0.53    6.3    1.11 10.28 15.33              4.23
      Source: ASI, CSO.



      Table A.13.2b: Shares of states in food manufacturing (2000–01 and 2005–06)
                                    AP    AS      BH          CH        GJ      HY   HP     JH       KT      KE         MP    MH     OR    PB      RJ     TN         UP     UK   WB
      Enterprise        2000-01    8.8    2.8     8.0     1.3           2.5    1.0   1.0    3.2     3.9      2.1     3.0      7.7   6.9    1.3    4.0     4.5     15.7     0.8   18.7
                        2005-06    6.1    4.2     7.9     0.7           3.6    1.3   1.2    3.4     4.6      3.7     3.4      7.0   10.2   1.2    4.5     4.2     16.6     0.6   13.6
      Workers          2000-01 10.4       2.3     6.4     1.0           2.0    0.9   0.6    2.6     5.6      2.6     6.9      3.3   7.1    1.3    3.0     5.2     15.5     0.7   19.8
                       2005-06 10.1       3.2     6.1     0.6           2.9    1.3   0.6    2.4     8.2      4.2     5.6      2.9   9.0    1.0    3.0     5.3     15.9     0.7   14.1
      Investment       2000-01     8.9    1.1     4.4     0.8           3.4    3.4   1.6    1.4     4.6      4.1     11.6     3.7   2.5    5.0    4.6     8.6     14.9     1.0   7.9
                       2005-06     8.5    1.4     4.2     0.6       3.6        4.7   1.1    1.1     6.3      7.3     11.1     3.2   1.9    3.5    4.1     9.1     14.9     1.6   7.7
      Gross value      2000-01 10.4       2.0     5.4     0.7       3.1        1.8   0.9    1.9     5.9    3.6       8.8      3.2   3.4    2.8    3.6     7.7     12.6     0.8   16.1
      added
                       2005-06     9.1    2.9     4.4     0.4       4.1        3.2   0.5    1.8     11.4   4.2       8.1      3.6   3.3    2.4    4.3     7.6     14.6     0.7   10.4
      Source:  NSS.
      Note:  Investment or capital in unorganized refers to market value of fixed assets (own+hired).




245
                                                                                                                                                                                        ﻿Annex 13: Food Processing: Structure and Performance
                      Location Quotient and concentration/diversification of investment in the organized food industry
      Table A.13.3a:  
                      (decadal averages)




246
                                                   Location Quotient                        Relative Diversity Index
      State                              1980-89       1990-99         2000-09    1980-89           1990-99            2000-09
      Andhra Pradesh                      1.13           1.20           1.63        1.36             0.63               0.20
      Assam                               5.19           5.56           1.85        0.04             0.03               0.15
      Bihar                               0.37           1.50           2.15        0.28             0.26               0.11
      Gujarat                             0.73           0.44           0.33        0.65             0.23               0.19
      Haryana                             0.70           0.93           0.98        0.57             1.73               5.90
      Himachal Pradesh                    0.12           0.83           0.39        0.20             0.77               0.21
      Karnataka                           1.64           1.24           0.90        0.27             0.53               1.26
      Kerala                              0.64           1.56           2.04        0.48             0.23               0.12
      Madhya Pradesh                      0.57           0.96           1.81        0.40             2.96               0.16
      Maharashtra                         1.14           1.03           0.96        1.27             4.77               3.46
      Odisha                              0.10           0.48           0.31        0.19             0.25               0.19
      Punjab                              1.53           2.31           2.12        0.33             0.10               0.11
      Rajasthan                           0.47           0.55           0.64        0.33             0.28               0.35
      Tamil Nadu                          0.85           1.14           0.73        1.14             0.95               0.48
                                                                                                                                 Republic of India: Accelerating Agricultural Productivity Growth




      Uttar Pradesh                       1.62           1.52           4.13        0.28             0.25               0.04
      West Bengal                         0.49           0.79           0.95        0.34             0.61               2.30
      Average                               –              –              –         0.51             0.86               0.95
      Source:  Bathla and Gautam 2013.
      Table A.13.3b:  Location Quotient and concentration/diversification of investment in the unorganized food industry

                                                 Location Quotient                                 Relative Diversity Index
                                     2000-01                         2005-06                   2000-01                   2005-06
      State                  Rural       Urban    Total    Rural     Urban     Total   Rural   Urban     Total   Rural   Urban      Total
      Andhra Pradesh         1.34        1.55      1.64    1.12       1.57      1.4    0.09     0.14     0.08    0.27     0.14      0.14
      Assam                  1.62        1.38      2.06    1.46       1.16     1.81    0.05     0.21     0.05    0.07     0.52      0.07
      Bihar                  1.72        1.94      2.34    1.73       2.45     2.55    0.04     0.08     0.04    0.04     0.06      0.04
      Chhattisgarh           1.05        1.35      1.28     0.4       0.42     0.44    0.68     0.23     0.19    0.05     0.14       0.1
      Gujarat                0.83        0.49       0.5    0.73       0.59     0.51    0.19     0.16     0.11    0.12         0.2   0.11
      Haryana                1.21        1.12      1.07    0.68       0.99     0.76    0.15     0.65     0.82     0.1    11.31      0.23
      Himachal Pradesh       1.51         1.7       2.2    1.09        1.6     1.59    0.07     0.13     0.05    0.35     0.14      0.09
      Jharkhand              0.74        2.23      1.28    0.93       1.61     1.38    0.12     0.06     0.19    0.47     0.13      0.14
      Karnataka              0.94          1       0.97    0.87       1.36      1.1    0.53    18.25     1.91    0.24     0.23      0.55
      Kerala                 0.72        1.58      1.16    1.07       1.18     1.45    0.11     0.13     0.33     0.5     0.46      0.12
      Maharashtra            0.94        0.86      0.77    1.04       0.72     0.67    0.55     0.55     0.23    0.77     0.29      0.17
      Madhya Pradesh         1.07         1.7      1.38    0.83       1.68     1.22    0.47     0.11     0.14    0.19     0.12      0.24
      Odisha                 1.45        1.87       2.1    1.15       1.89     1.68    0.07     0.09     0.05    0.22     0.09      0.08
      Punjab                 0.67        1.22       0.9    0.95       0.93     0.84     0.1     0.36     0.54    0.69     1.15      0.33
      Rajasthan              0.91        1.24      1.11     0.9          1     0.95    0.37     0.33     0.49    0.31    16.94         1
      Tamil Nadu             0.51        1.02      0.74    0.57        1.1     0.81    0.07      3.2     0.21    0.07     0.85      0.29
      Uttar Pradesh          1.13        1.31      1.37    1.43       1.11     1.36    0.25     0.25     0.14    0.07     0.74      0.15
      Uttaranchal            1.33        1.53       1.7    1.14       3.34     2.08     0.1     0.15     0.08    0.23     0.03      0.05
       West Bengal           0.97        1.21      1.19    1.04       1.01     1.18    0.97     0.37     0.28    0.88    10.47       0.3
       Average                                                                         0.28     1.28      0.3    0.29     2.21      0.22
      Source:  Bathla and Gautam 2013.




247
                                                                                                                                            ﻿Annex 13: Food Processing: Structure and Performance
                     Descriptive statistics on private investment in organized food manufacturing (average 1985–2009; values
      Table A.13.4:  
                     in 2004–05 price)




248
                                                      Rate of
                                                      Return:                Change in    Capacity
                            Investment/ Investment/    Profit/   Output/      Output/    Utilization:   Cashflow/   Outstanding
      State                   Capital   Value Added   Capital    Capital      Capital    Capital/VA      Capital    Loan/Capital
      Andhra Pradesh            0.12       0.32        0.09        3.51         0.22        2.73          0.29          0.55
      Assam                     0.11       0.25        0.27        2.22         0.01        2.76          0.29          0.36
      Bihar                     0.10       0.32        0.07        2.45         0.05        3.19          0.24          0.69
      Gujarat                   0.12       0.34        0.05        4.36         0.39        2.91          0.39          0.64
      Haryana                   0.15       0.30        0.17        4.36         0.29        2.05          0.55          0.53
      Himachal Pradesh          0.17       0.62        0.06        3.55         0.13        3.29          0.23          0.45
      Karnataka                 0.14       0.36        0.10        2.49         0.18        2.59          0.27          0.62
      Kerala                    0.14       0.19        0.33        5.62         0.32        1.44          0.44          0.58
      Madhya Pradesh            0.12       0.22        0.14        4.68         0.48        2.83          0.42          0.73
      Maharashtra               0.12       0.33        0.06        2.79         0.16        2.74          0.26          0.64
      Odisha                    0.15       0.54        0.03        3.39         0.25        3.66          0.27          0.67
      Punjab                    0.13       0.25        0.26        3.76         0.20        1.87          0.30          0.45
      Rajasthan                 0.12       0.33        0.14        4.36         0.31        2.78          0.33          0.59
      Tamil Nadu                0.13       0.30        0.15        2.93         0.18        2.45          0.28          0.45
                                                                                                                                   Republic of India: Accelerating Agricultural Productivity Growth




      Uttar Pradesh             0.14       0.42        0.05        2.43         0.17        3.20          0.18          0.55
      West Bengal               0.13       0.40        0.06        3.63         0.09        3.13          0.31          0.47
      Source:  Bathla and Gautam 2013.
                                                               ﻿Annex 13: Food Processing: Structure and Performance




Table A.13.5:  Determinants of Investment in Food Processing

                                                                             Gross Fixed Investment/Capital Stock (-1)
 Lagged Dependent Variable                                                          -0.289                        -0.262
 ((Gross fixed investment/capital stock (-1)(-1))                                 (5.96)***                     (5.45)***
 Changes in Demand & Adjustment of Capital                                           0.015                        0.017
 (Change in output/capital stock (-1))                                            (4.56)***                     (4.98)***
 Accelerator or Expected Profit                                                      0.026                        0.043
 (Output (-1)/capital stock)                                                      (2.88)***                     (4.14)***
 Availability of Credit                                                             -0.100                        -0.019
 (Loan outstanding/capital stock) (-1)                                              (1.75)*                       (0.37)
 Infrastructure                                                                      0.534                        0.172
 (Capital expend. on roads & bridges per capita, lagged)                           (2.57)**                       (0.89)
 Agriculture Linkage                                                                 0.005
 (Agriculture SDP/SDP) (-1)                                                       (2.85)***                            --
 Land Productivity                                                                                                0.266
 (Agriculture SDP/NSA)                                                                    --                     (2.17)**
 Localization Index                                                                  0.044                        0.026
 (Location quotient of factories)                                                    (0.78)                       (0.46)
 Investor Friendliness                                                              -0.101                        -0.011
 (No. strikes or lockouts per factory)                                               (0.40)                       (0.04)
 Constant                                                                           -0.066                        -0.054
                                                                                     (0.48)                       (0.39)
 State effects                                                                         Yes                           Yes
 R2                                                                                   0.23                          0.22
 N                                                                                     374                          374
Source:  Bathla and Gautam 2013.
Note:  Estimated coefficients are elasticities. t-statistics in parentheses. Level of significance: ***: 1%; **: 5%; *: 10%.




                                                                                                                               249
                     Annual rate of growth in partial productivity in organized and unorganized manufacturing
      Table A.13.6:  
                     (2004–05 prices)




250
                                                            Organized Sector                                            Unorganized Sector
                                                                                                                      Labor     Capital    Capital
                           Labor Productivity               Capital Intensity             Capital Productivity        Prod.      Int.       Prod.
      State            1980-89 1990-99 2000-08 1980-89 1990-99 2000-08 1980-89 1990-99 2000-08 2000-05 2000-05 2000-05
      AP                   11.41      4.54      4.39       9.01     10.04        4.71      2.21       -4.99   -0.30     3.64       3.56       0.08
      Assam                11.57     -0.57      4.09     16.54       6.36        2.39     -4.27       -6.51   1.66      6.92       2.60       4.21
      Bihar                18.65    12.03       1.37     32.09      14.19        2.78    -10.18       -1.89   -1.37     2.40       3.55       -1.11
      Gujarat               9.14      3.45      3.62     12.90       8.57        1.95     -3.34       -4.72   1.64      3.80       -2.56      6.52
      Haryana              11.04      7.63      9.06     10.28      13.06       10.33      0.68       -4.80   -1.15    11.65       3.60       7.77
      HP                    5.37      9.11     -1.00       5.37     12.23        0.33    -0.003       -2.78   -1.33     -6.86      -3.19      -3.79
      Karnataka            14.24      6.84      8.58     17.91       9.67        4.69     -3.11       -2.58   3.71     11.47       2.52       8.73
      Kerala               13.97      4.52     -0.77     17.85      11.31        2.85     -3.29       -6.09   -3.52     -1.02      5.45       -6.14
      MP                   17.38      5.54      9.93     28.60      16.75        1.41     -8.73       -9.61   8.40     11.60       3.43       7.90
      Maharashtra          14.80      6.74      3.25     16.69       6.64        4.55     -1.61       0.09    -1.24     8.29       7.18       1.03
      Odisha                3.46      2.77      8.29     12.70      12.48        8.03     -8.20       -8.63   0.24      0.47       -5.90      6.77
      Punjab                4.43      3.95      9.93     11.46       6.37        5.71     -6.31       -2.27   3.99      7.51       1.30       6.13
      Rajasthan             8.37      7.18      9.41       9.91      9.82        0.12     -1.40       -2.40   9.28      9.59       1.68       7.79
                                                                                                                                                      Republic of India: Accelerating Agricultural Productivity Growth




      Tamil Nadu           11.62      2.28      5.28     14.84       7.45        3.42     -2.81       -4.82   1.80      4.98       4.52       0.45
      UP                   12.72      8.35      4.63     18.81      12.19        9.15     -5.13       -3.42   -4.14     8.33       3.32       4.85
      WB                   10.27      1.20      9.44     13.15       9.56        3.46     -2.54       -7.63   5.77      3.59      10.69       -6.42
      India                12.15      5.19      5.87     16.76       8.71        4.87     -3.95       -3.23   0.96      5.75       3.87       1.81
      Source:  Bathla and Gautam 2013.
      Note: Capital for unorganized sector refers to fair market value of fixed assest (own+hired).
                                                   ﻿Annex 13: Food Processing: Structure and Performance




Table A.13.7:  Total factor productivity growth in food manufacturing (2004–05 prices)
                                     Organized sector                             Unorganized Sector
                          TFP Index                    TFP Growth                    Annual Growth
                      (Decadal Average)             (Percent Per Year)             (2000-01 – 2005-06)
State            1980-89 1990-99 2000-09 1980-89 1990-99 2000-09              Rural      Urban      Total
AP                  88        96       94       2.6        -0.7         0.4     0.8       -6.7       -1.8
Assam             -138     -4.44       94         –           –        -8.8     1.0        3.5         1.4
Bihar              114      118      115        4.0         1.9        -0.2    -1.7        0.3       -1.3
Gujarat             97        94       99       1.1        -1.1         1.6     4.4        2.1         2.8
Haryana            125      115      109        3.3        -1.3         1.4     4.9        1.4         2.4
HP                 101        87     128          –         6.3        -2.8    -0.8       -3.7       -1.1
Karnataka          102        98     103        1.7         0.3         2.7     6.2       -3.8         2.4
Kerala             328      239      132        2.0        -2.2        -3.7    -4.7        2.7       -2.9
MP                 135      121      131       -0.2        -4.8         4.6     5.4       -0.6         2.8
Maharashtra        116      122      110        3.4         1.3         0.1    -0.1        1.4         0.6
Odisha             250       211     159       -3.0        -7.0         2.8     5.2        2.0         4.1
Punjab             167       142     143       -2.8        -1.1         4.2    -0.4        3.7         2.5
Rajasthan           66        92     105        2.4        -0.1         5.5     4.1        1.5         3.1
Tamil Nadu         154       128     109        0.8        -2.2         2.3    -2.4        0.5       -0.5
UP                 104       100       89       2.0         0.4        -1.5     2.8       -1.1         1.5
WB                 128       119       91       3.2        -2.7         6.4    -6.3       -0.9       -4.5
India              126       116     106        1.3        -0.5         1.8     0.3       -0.5         0.1
Source:  Bathla and Gautam 2013.


               Decadal averages of technical efficiency using time varying decay inefficiency model
Table A.13.8:  
Period                       AP          Assam           Bihar       Gujarat        Haryana          HP
1985-86 to 1989-90          0.77          0.79           0.87         0.85           0.97           0.96
1990-91 to 1999-00          0.70          0.72           0.82         0.80           0.96           0.95
2000-01 to 2008-09          0.59          0.62           0.75         0.72           0.94           0.93
1985-86 to 2008-09          0.67          0.70           0.80         0.78           0.96           0.94
                         Karnataka       Kerala           MP       Maharashtra      Odisha           PB
1985-86 to 1989-90          0.93          0.81           0.97         0.90           0.72           0.98
1990-91 to 1999-00          0.90          0.75           0.96         0.87           0.64           0.98
2000-01 to 2008-09          0.86          0.66           0.94         0.81           0.51           0.96
1985-86 to 2008-09          0.89          0.73           0.95         0.85           0.61           0.97
                         Rajasthan        TN              UP          WB
1985-86 to 1989-90          0.98          0.84           0.78         0.75
1990-91 to 1999-00          0.97          0.78           0.71         0.67
2000-01 to 2008-09          0.95          0.69           0.60         0.55
1985-86 to 2008-09          0.96          0.76           0.68         0.64
Source:  Bathla and Gautam 2013.




                                                                                                       251
Republic of India: Accelerating Agricultural Productivity Growth




Table A.13.9:  Determinants of TFP in organized food-processing

                                                                                        TFP Index
 Road density                                                                  0.030                 0.155
 (Km/’000 km )   2
                                                                               (0.87)               (1.68)*
 Agriculture electricity consumption                                           -0.168                0.033
 (% of consumption in total consumption)                                  (5.49)***                  (0.62)
 Total electricity consumption                                                 -0.025                -0.008
 (Million kilowatts per capita)                                                (0.64)                (0.10)
 Agricultural linkage                                                          0.335                 0.290
 (Agricultural SDP/total SDP)                                             (3.05)***                 (2.07)**
 Urbanization                                                                  0.582                 0.022
 (% urban population)                                                     (5.50)***                  (0.07)
 Localization index                                                            -0.060                0.132
 (Location Quotient based on factories)                                        (1.08)                (1.05)
 Investor friendliness                                                         -0.047                -0.017
 (No. of strikes/lockouts per factory)                                    (6.31)***                 (2.54)**
 Trend                                                                         -0.018                -0.001
                                                                           (2.21)**                  (0.12)
 Year effects                                                                    Yes                    Yes
 State effects                                                                    No                    Yes
 Constant                                                                      37.56                   4.24
                                                                           (2.27)**                  (0.23)
 Adj. R2                                                                       0.188                 0.525
 N                                                                               381                   381
Source:  Batha and Gautam 2013.
Note:  Estimated coefficients are elasticities. t-statistics in parentheses.
Level of significance: ***: 1% ; **: 5%; *: 10%.




  252
                              Annex 14: Bihar and Odisha:
annexes                       Agriculture Performance and
                              Constraints



                 verage annual growth rates in AGDP and SGDP for selected states (2004–12)
 Figure A.14.1: A

 14.0%


 12.0%


 10.0%


  8.0%


  6.0%


  4.0%


  2.0%


  0.0%
          Odisha


                   Bihar


                           Chhattisgarh


                                          MP


                                                Jharkhand


                                                            UP

                                                                 Rajasthan




                                                                               Maharashtra


                                                                                             Gujarat


                                                                                                       West Bengal


                                                                                                                      Karnataka


                                                                                                                                  Tamil Nadu


                                                                                                                                               Uttarakhand


                                                                                                                                                               AP


                                                                                                                                                                        HP

                                                                                                                                                                             Haryana


                                                                                                                                                                                       Punjab

                                                                                                                                                                                                Kerala
  -2.0%




                                               LIS                                                                                                Non-Lis

                                                     AGDP Growth             SGDP Growth                             AGDP Avg                                SGDP Avg


 Source:  Authors, using CSO data.




                                                                                                                                                                                                253
Republic of India: Accelerating Agricultural Productivity Growth




               Poverty rates in LIS compared to selected non-LIS (2004–05, 2011–12)
Figure A.14.2: 

   70

   60

   50

   40

   30

   20

   10

    0
                                                            Madhya Pradesh
            Odisha

                          Bihar


                                           Chhattisgarh




                                                                               Jharkhand


                                                                                                      UP

                                                                                                                    Rajasthan




                                                                                                                                       Maharashtra


                                                                                                                                                            Gujarat

                                                                                                                                                                         West Bengal


                                                                                                                                                                                           Karnataka


                                                                                                                                                                                                                Tamil Nadu

                                                                                                                                                                                                                                         Uttarakhand


                                                                                                                                                                                                                                                                 Andhra Pradesh

                                                                                                                                                                                                                                                                                                Himachal Pradesh


                                                                                                                                                                                                                                                                                                                                      Haryana


                                                                                                                                                                                                                                                                                                                                                                 Punjab

                                                                                                                                                                                                                                                                                                                                                                               Kerala
                                                                             LIS                                                                                                                                               Non-Lis

                                                                              Rural HCR 04-05                                    Rural HCR 11-12                                          Avg 04-05                                              Avg 11-12



Source:  Calculated using estimates from Planning Commission as follows: 1993–94 and 2004–05 Estimates: Press Note on
Poverty Estimates, Planning Commission, Govt. of India, Jan 2011; 2009–10 Estimates: Press Note on Poverty Estimates,
2009-10, Planning Commission, Govt. of India March 2012; 2011–12 Estimates: Press Note on Poverty Estimates, 2011–12,
Planning Commission, GOI, July 2013.
Note:  The estimates for Chhattisgarh, Madhya Pradesh, Bihar, Jharkhand, Uttar Pradesh, and Uttaranchal are for states as
they exist after bifurcation in 2001. The estimates for 1993–94 have been calculated from the unit data using district and
state boundaries of the divided states in 1993–94. Poverty estimates are based on the new official poverty line, using the
Tendulkar methodology. Population as on 1st March 2010 has been used for estimating number of persons below poverty
line (interpolated between 2001 and 2011 population census).


               Rural poverty rates and share of agriculture in state GDP (2004–05, 2011–12)
Figure A.14.3: 
    0.3                                                                                                                                                                                                                                                                                                                                                                             70
                                                                                                                                                                                                                                                                                                                                                                                    60
   0.25
                                                                                                                                                                                                                                                                                                                                                                                    50
    0.2
                                                                                                                                                                                                                                                                                                                                                                                    40
   0.15                                                                                                                                                                                                                                                                                                                                                                             30
    0.1                                                                                                                                                                                                                                                                                                                                                                             20
   0.05                                                                                                                                                                                                                                                                                                                                                                             10
        0                                                                                                                                                                                                                                                                                                                                                                               0
                                                                                   Madhya Pradesh
              Jharkhand

                            Chhattisgarh

                                              Odisha

                                                             Rajasthan




                                                                                                    Uttar Pradesh

                                                                                                                    Bihar




                                                                                                                                   Maharashtra

                                                                                                                                                     Tamil Nadu

                                                                                                                                                                      Kerala

                                                                                                                                                                                       Gujarat

                                                                                                                                                                                                       Uttarakhand

                                                                                                                                                                                                                             Karnataka

                                                                                                                                                                                                                                                       Haryana


                                                                                                                                                                                                                                                                                  West Bengal

                                                                                                                                                                                                                                                                                                                   Himachal Pradesh

                                                                                                                                                                                                                                                                                                                                                Andhra Pradesh

                                                                                                                                                                                                                                                                                                                                                                      Punjab




                                                            LIS                                                                                                                                        Non-Lis
                                                          Rural HCF 2004/05                                               Rural HCR 2011/12                              Ag Share of GSDP (Average 2005-12)
Source:  Authors, using data from Planning Commission website and CSO.



  254
                                          ﻿Annex 14: Bihar and Odisha: Agriculture Performance and Constraints




Table A.14.1: Sector shares in Bihar, Odisha, and India overall between 2004 and 2012 (%)
               2004–05        2005–06       2006–07       2007–08       2008–09       2009–10        2010–11     2011–12
Bihar
Agric.             32            30            31            28             27              24          23          22
Industry           14            15            15            17             18              19          19          20
Services           55            55            53            55             55              58          58          58
Odisha
Agric.             23            23            21            20             19              19          18          16
Industry           34            33            36            38             37              35          35          36
Services           42            44            44            43             45              46          47          48
India
Agric.             19            18            17            17             16              15          14          14
Industry           18            18            19            19             18              18          18          18
Services           63            64            64            65             66              67          67          68
Source:  Estimates from MOSPI 2013.
Note:  Agriculture and allied sector includes agriculture, forestry, and fishing.

Table A.14.2: Distribution of land holdings in Bihar (2005–06) and Odisha (2006–07)
Holding size (ha)                               Number of holdings (million)                 Share of total holdings (%)
Bihar
<0.5                                                           10.6                                    72.3
0.5–1.0                                                         2.5                                    17.3
1.0–2.0                                                           1                                     6.7
2.0–5.0                                                         0.5                                     3.4
> 5.0                                                          0.04                                     0.3
Odisha
<1.0                                                            2.6                                    59.6
1.0–2.0                                                        1.16                                    26.5
2.0–4.0                                                        0.47                                    10.8
4.0–10.0                                                       0.12                                     2.7
> 10.0                                                         0.01                                     0.3
Source: MOA 2005.

              Value of output of major commodities in total value of output of agriculture,
Table A.14.3: 
              Bihar and Odisha (%)
                                                 Bihar                                               Odisha
                               1990–91          2001–02           2008–09           1990–91         2001–02        2008–09
Rice                              20               13                11              29.7             29.3          24.5
Wheat                             14               11                 8
Maize                              3                3                 2               0.7              0.1            0.3
Pulses                             7                3                 2              11.9              2.4            2.9
Fruits and Vegetables             14               28                23              22.1             31.2           29.4
Livestock                         30               25                40               9.2             11.8           18.2
Fisheries                                           3                 4                                5.8            6.6
Source: MOSPI 2013.




                                                                                                                      255
      TableA.14.4:  Biotic problems by physiographic zone in Odisha




256
                             Zone I                        Zone II                        Zone III                      Zone IV
                       (Central table land)             (Eastern ghat)                (Coastal plains)            (Northern plateau)     Extent of damage
      Crops         Diseases          Pest         Diseases          Pest        Diseases           Pest        Diseases        Pest     caused ( per cent)
      Paddy       Blast        Stem-borer,       Blast,        BPH, Stem-     Blast (every    Sucking         Blast        Stem-borer Zone I: Blast
                               swarming         (fungal        borer (once    year, rampant pests; green                                 (20%), stem borer
                               caterpillar      diseases)      in a season)   during          leafhopper                                 (20%), swarming
                                                (once in a                    September       (every year,                               caterpillar (25%);
                                                season)                       & October),     rampant                                    Zone II: Blast
                                                                              bacterial       during                                     (25%), BPH (25%),
                                                                              blight          September &                                Stem-borer
                                                                                              October)                                   (25%); (Zone III:
                                                                                                                                         Sucking pests
                                                                                                                                         (50%), green
                                                                                                                                         leafhopper (50%),
                                                                                                                                         Blast (50%);
      Maize       Turcicum      Stem-borer,     Turcicum      Stem-borer,     Turcicum leaf     Stem-borer, Turcicum       Stem-borer, Turcicum leaf
                  leaf blight   swarming        leaf blight   aphid,          blight            aphid,        leaf blight aphid,         blight (up to 70%)
                                caterpillar                   defoliating                       defoliating                defoliating
                                                              caterpillars,                     caterpillars,              caterpillars,
                                                              termite                           termite,                   termite,
      Wheat      Loose smut,    Termite, pink Loose smut, Termite, pink       Loose smut,       Termite, pink Loose smut, Termite,
                 leaf blight,   borer, aphid     leaf blight, borer, aphid    leaf blight,      borer, aphid leaf blight, pink borer,
                 alternaria                      alternaria                   alternaria leaf                 alternaria aphid
                 leaf blight,                    leaf blight,                 blight, wilt                    leaf blight,
                 wilt                            wilt                                                         wilt
      Pulses     Powdery        Leaf eating      mildew,      Leaf eating     mildew, seed      Leaf eating   mildew,      Leaf eating
                 mildew,        caterpillars and seed rot & caterpillars      rot & seedling    caterpillars seed rot & caterpillars
                                                                                                                                                              Republic of India: Accelerating Agricultural Productivity Growth




                 seed rot &     pod borers,      seedling     and pod         damage, wilt,     and pod       seedling     and pod
                 seedling       aphids, white damage,         borers,         collar rot,       borers,       damage,      borers,
                 damage,        fly              wilt, collar aphids, white   yellow mosaic     aphids, white wilt, collar aphids,
                 wilt, collar                    rot, yellow fly              virus, leaf       fly           rot, yellow white fly
                 rot, yellow                     mosaic                       spot                            mosaic
                 mosaic                          virus, leaf                                                  virus, leaf
                 virus, leaf                     spot                                                         spot
                 spot
      Vegetables Blight;      Stem-borer,       Blight;      Stem-borer,      Blight;         Stem-borer,     Blight;      Stem-borer,
                 damping-off aphids and         damping-off aphids and        damping-off aphids and          damping-off aphids and
                 of seedlings jassids           of seedlings jassids          of seedlings jassids            of seedlings jassids
                 in brinjal,                    in brinjal,                   in brinjal,                     in brinjal,
                 tomato                         tomato                        tomato and                      tomato
                 and chilli;                    and chilli;                   chilli; leaf                    and chilli;
                 leaf spot,                     leaf spot,                    spot, bacterial                 leaf spot,
                 bacterial                      bacterial                     and fungal                      bacterial
                 and fungal                     and fungal                    wilt; mildew                    and fungal
                 wilt;                          wilt; mildew                                                  wilt; mildew
                 mildew
      Source:  IFPRI 2012b.
      Table A.14.5: Abiotic and institutional problems by physiographic zone in Odisha
                                      Zone I
      Sl.                          (Central table              Zone II                        Zone III                     Zone IV)
      No.     Type of problem          land)                (Eastern ghat)                 (Coastal plains)           (Northern plateau)
      1.     Flood               Every year           No                            Every year (also cyclone-      No
                                                                                    prone and hail-prone)
      2.     Drought             Slightly drought     Once in 2 to 3 years          No                             Drought prone (low
                                 prone                                                                             rainfall)
      3.     Salinity/alkalinity Slight alkalinity    No                            Yes                            Salinity and alkalinity
                                 in some parts
      4.     Acidity             Exists in some       Exists in some parts          Moderately acidic              Highly acidic
                                 parts
      5      Water-logging       Every year           Every year                    8 to 10 days in 5 to 6 years   No
      6.     Credit              Poor access to       Poor access to institutional Short-term loan (for 6           Poor access to
                                 institutional loan   loan                         months) is available             institutional loan
                                                                                   through primary
                                                                                   cooperative society (interest
                                                                                   rate 9% p.a.), but it is grossly
                                                                                   insufficient for farmers who
                                                                                   take easily-accessible loan
                                                                                   from local money-lender
                                                                                   (interest rate 36% p.a.)
      7.     Marketing           Farmers sell         Farmers sell the produce      Farmers sell the produce       Farmers sell the
                                 the produce to       to traders due to lack of     to local traders due to lack   produce to traders
                                 traders due to       effective procurement         of effective procurement       due to lack of effective
                                 lack of effective    by government agency;         by government agency;          procurement by
                                 procurement          farmers bear the              farmers bear the transport-    government agency.
                                 by government        transport-charges and         charges and receive a price
                                 agency.              receive a price of Rs 1,000   of Rs 750 to 850 per quintal
                                                      per quintal against MSP of    against MSP of
                                                      Rs 1,080 per quintal          Rs 1080 per quintal
      Source:  IFPRI (2012b).




257
                                                                                                                                              ﻿Annex 14: Bihar and Odisha: Agriculture Performance and Constraints
Republic of India: Accelerating Agricultural Productivity Growth




                griculture share of electricity consumption and irrigation intensity in selected states (%)
Figure A.14.4: A
                                                                                                                                                                          98
100
90                                                                                                                                                         84
80
70                                  62
                                                                                                59
60
50                                                      44                                                             43                                                                    45
                                                                                                                                       43
                                                                        38                                                                   38
40
                                         29                                                          31                                                         32
                                                                                                                            26
30
                                                                                                                                                                                    20
                          16
20
                      5
10       1                                                    1              3
 0
                                          Madhya Pradesh*
             Assam*




                           Bihar*




                                                                  Odisha*




                                                                                 West Bengal*




                                                                                                      Andhra Pradesh




                                                                                                                             Gujarat




                                                                                                                                                 Haryana




                                                                                                                                                                 Punjab




                                                                                                                                                                                     India
                                          Agriculture Share of Electricity Consumption                                      Irrigation Intensity

Source:  Planning Commission website.
Note:  * indicates status as lagging state; irrigation intensity is irrigated area as share of net sown area in 2007.




Table A.14.6: Spending and growth in spending on agricultural research and extension
                  Spending
                (Rs millions)                                                        Average annual growth in spending (%)
                     2007                                    1971–1980                   1981–1990         1991–2000                                                       2001-08
Agricultural research
Bihar                289.76                                        27                                  9                                     0                                 11
Odisha               169.11                                        16                                  7                                    –1                                 15
India              6,244.22                                         9                                  8                                     7                                  2
Agricultural extension
Bihar                 97.27                                        –7                                 8                                     –6                                  0
Odisha                11.34                                        –3                                18                                      7                                  2
India              1,654.81                                        –3                                11                                      3                                  7
Source:  IFPRI 2012a and 2012b.




  258